WO2011111284A1 - 関連情報出力装置、関連情報出力方法およびコンピュータ読み取り可能な記録媒体 - Google Patents
関連情報出力装置、関連情報出力方法およびコンピュータ読み取り可能な記録媒体 Download PDFInfo
- Publication number
- WO2011111284A1 WO2011111284A1 PCT/JP2010/073724 JP2010073724W WO2011111284A1 WO 2011111284 A1 WO2011111284 A1 WO 2011111284A1 JP 2010073724 W JP2010073724 W JP 2010073724W WO 2011111284 A1 WO2011111284 A1 WO 2011111284A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- information
- recognition
- degree
- destination information
- source
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/955—Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]
- G06F16/9558—Details of hyperlinks; Management of linked annotations
Definitions
- the present invention relates to a related information output device, a related information output method, and a computer-readable recording medium, and more particularly, to a related information output device that outputs unexpected information that attracts the user's interest from related source information, and related information output.
- the present invention relates to a method and a computer-readable recording medium.
- the related information output device is a device that outputs information related to related source information (hereinafter also referred to as related information).
- related information includes a combination of related source information and related destination information (destination information), or a combination of related source information, related destination information, and related content (relation content). That is, the related information is information indicating a relationship between a keyword, a document, an image, and sound.
- the related content is a keyword and a sentence indicating the type and characteristics of the relationship.
- Non-Patent Document 1 An example of a conventional related information output device is Shingo Otsuka et al., “A Study on Discovering Related Words Using Global Web Access Logs”, IPSJ Journal, vol. 46, No. SIG8 (TOD26), pp. 82-92 (2005) (Non-Patent Document 1).
- related source information and related destination information are keywords.
- the keyword of the related source is used as an input, and the keyword with the highest frequency is output among the keywords existing in the Web page browsed so far. Therefore, the related information output device described in Non-Patent Document 1 is intended to present a keyword most relevant to related source information.
- Non-Patent Document 2 discloses a method for evaluating whether or not the information is surprising based on the co-occurrence frequencies of two categories.
- information belongs to a plurality of categories, and the co-occurrence frequency of two categories is the number of times two categories belong to one information at the same time. It is assumed that the information Z belonging to two categories having a low frequency of co-occurrence of categories includes unexpected knowledge that straddles these two categories. When the information Z is included in any two categories having a low co-occurrence frequency, the information Z is output as unexpected information.
- the information “Taro Aso” belongs to the category “Prime Minister” and the category “Olympic Player”. At this time, when the co-occurrence frequency of these two categories is low, the information “Taro Aso” is output as information that is surprising.
- Non-Patent Document 1 The related information output device described in Non-Patent Document 1 is made for the purpose of presenting related destination information that is most related to related source information. For this reason, there is a problem that the click rate on the website is low because information that the user knows (known information) is presented and related information that attracts the user's interest cannot be presented. . In addition, even if the judgment method is reversed and information that is not related is presented, the information that is not known is not selected by the unknown related information that no one knows. There was a problem that the rate was low.
- Non-Patent Document 2 alone, it is not possible to present unexpected related information that attracts the user's interest. This is because simply because it belongs to a category with low co-occurrence frequency cannot be said to be surprising information.
- Non-Patent Document 2 since information with low co-occurrence frequency is presented, even if related source information itself and related destination information itself do not appear so much, they are output. When the related source information itself and the related destination information itself are unknown information, the output information becomes information that is completely uninteresting for the user, so the user does not feel any surprise.
- the present invention has been made to solve the above-described problems, and an object of the present invention is to provide a related information output device, a related information output method, and a computer reading that can output unexpected information that attracts the user's interest. It is to provide a possible recording medium.
- a related information output device for outputting information related to related source information, and receives the related source information, A plurality of related destination information related to related source information, an appearance frequency of the related source information, an appearance frequency of each related destination information, an appearance frequency of the relationship between the related source information and each related destination information Is acquired from the related information storage unit provided inside or outside the related information output device, and the degree of recognition of the related source information is acquired based on each acquired appearance frequency.
- the degree of recognition of each related destination information and the degree of recognition for calculating the degree of recognition of each of the above relationships, the degree of recognition of the related source information, the degree of recognition of the related destination information, and the degree of recognition of the relation 2 or more pairs A high interest information narrowing section for selecting one or a plurality of the related destination information from the related destination information based on the matching, and outputting at least the selected related destination information.
- a related information output method is a related information output method in a related information output apparatus for outputting information related to related source information, the related source information
- the plurality of related destination information related to the related source information the appearance frequency of the related source information, the appearance frequency of each related destination information, the related source information and the related destination information
- the step of acquiring the appearance frequency of the relationship from the related information storage unit provided inside or outside the related information output device, and the degree of recognition of the related source information is acquired based on each acquired appearance frequency
- Two or more of the step of calculating the degree of recognition of each of the related destination information and the degree of recognition of each of the relations, the degree of recognition of the related source information, the degree of recognition of the related destination information, and the degree of recognition of the relation Based on the combination selects one or more of the above relation destination information from the respective associated destination information, and a step of at least outputting the relevant destination information selected.
- a computer-readable recording medium records a related information output program used in a related information output device for outputting information related to related source information.
- a computer-readable recording medium that receives the related source information, a plurality of related destination information related to the related source information, an appearance frequency of the related source information, and an appearance of each related destination information Acquiring the frequency and the appearance frequency of the relationship between the related source information and each related destination information from a related information storage unit provided inside or outside the related information output device, and each acquired appearance frequency And calculating the degree of recognition of the related source information, the degree of recognition of each acquired related destination information, and the degree of recognition of each of the relations, and One or a plurality of the related destination information is selected from the related destination information based on a combination of two or more of the recognition degree of the federation information, the recognition degree of the related destination information, and the recognition degree of the relation.
- a related information output program for causing the computer to execute at least the step of outputting the selected related destination information is
- FIG. 1 is a schematic configuration diagram of a related information output device according to a first embodiment of the present invention.
- FIG. 3 is a block diagram showing a control structure provided by the related information output device according to the first embodiment of the present invention.
- 5 is a flowchart showing an operation procedure when the related information output device according to the first embodiment of the present invention performs related information output processing.
- FIG. 5 is a diagram showing an example of accumulated information in a related information accumulation unit in the related information output device according to the first embodiment of the present invention.
- 6 is a diagram illustrating an example of a rule for unexpectedness determination in a high interest information narrowing-down unit 54.
- FIG. FIG. 10 is a block diagram showing a control structure provided by a related information output device according to a second embodiment of the present invention.
- FIG. 10 is a flowchart showing an operation procedure when the related information output apparatus according to the second embodiment of the present invention performs related information output processing.
- FIG. 10 is a diagram illustrating an example of a rule for unexpectedness determination in the high interest information narrowing unit 64.
- FIG. 10 is a block diagram showing a control structure provided by a position information display device according to a third embodiment of the present invention.
- FIG. 10 is a diagram showing an example of input content provided by an input unit in a position information display device according to a third embodiment of the present invention.
- FIG. 12 is a diagram showing an example of accumulated information in a position information accumulating unit in a position information display device according to a third embodiment of the present invention.
- FIG. 10 is a diagram showing an example of output contents of a position information display device according to a third embodiment of the present invention.
- the relation destination information is narrowed down. It is considered that highly interesting and interesting information can be presented by presenting to the user.
- the existence of a relationship is referred to as “link” or “relation”.
- link between the related source information “Olympic athlete” and the related destination information “Prime Minister” means that there is a link between “Olympic athlete” and “Prime Minister”.
- Non-Patent Documents 1 and 2 described above the major degree and minor degree of the combination of two or more of the related source information, the related destination information, and the link are not considered, so the user is interested. There was a problem that unexpected information could not be presented.
- the related information output apparatus receives one or more related source information, and relates to related source information from a specified database such as Web and Wikipedia (registered trademark).
- a related information search unit for searching for related information, a degree of recognition of the related source information searched, that is, a degree of recognition (recognition degree), a degree of recognition of related destination information, and a degree of recognition of a link;
- a high interest information narrowing section that narrows down unexpectedly interesting information based on the degree of recognition of related information.
- the awareness calculation unit calculates the related source information, the related destination information, and the link recognition level, respectively, and the high interest information screening unit is interested in the user based on a combination of two or more of these recognition levels Resolve issues by narrowing down to unexpected information.
- the related information output apparatus typically has a basic structure of a computer having a general-purpose architecture, and will be described later by executing a preinstalled program.
- Various functions are provided.
- a program is stored in a recording medium such as a flexible disk (Flexible Disk) and a CD-ROM (Compact Disk Read Only Memory) or distributed via a network or the like.
- an OS Operating System
- an OS for providing the basic function of the computer. System
- System may be installed.
- the program according to the first embodiment of the present invention executes processing by calling necessary modules out of program modules provided as a part of the OS in a predetermined order and / or timing. It may be a thing. That is, the program itself according to the first embodiment of the present invention does not include the module as described above, and the process may be executed in cooperation with the OS. Therefore, the program according to the first embodiment of the present invention may have a form that does not include the above-described module.
- the program according to the first embodiment of the present invention may be provided by being incorporated in a part of another program such as an OS. Even in this case, the program itself according to the first embodiment of the present invention does not include a module included in another program as described above, and the process is executed in cooperation with the other program. Is done. That is, the program according to the first embodiment of the present invention may be a form incorporated in such other programs.
- program execution may be implemented as a dedicated hardware circuit.
- FIG. 1 is a schematic configuration diagram of a related information output apparatus according to the first embodiment of the present invention.
- related information output device 201 is an information processing device such as a portable information terminal, a personal computer, and a server.
- CPU (Central Processing Unit) 101 that is an arithmetic processing unit, main memory 102, and hard disk 103
- An input interface 104, a display controller 105, a data reader / writer 106, and a communication interface 107 are connected to each other via a bus 121 so that data communication is possible.
- the CPU 101 performs various operations by developing programs (codes) stored in the hard disk 103 in the main memory 102 and executing them in a predetermined order.
- the main memory 102 is typically a volatile storage device such as DRAM (Dynamic Random Access Memory), and stores data indicating various arithmetic processing results in addition to a program read from the hard disk 103. To do.
- the hard disk 103 is a nonvolatile magnetic storage device, and stores various setting values in addition to programs executed by the CPU 101.
- the program installed in the hard disk 103 is distributed in a state of being stored in the recording medium 111, as will be described later.
- a semiconductor storage device such as a flash memory may be employed.
- the input interface 104 mediates data transmission between the CPU 101 and an input unit such as a keyboard 108, a mouse 109, and a touch panel (not shown). That is, the input interface 104 receives an external input such as an operation command given by the user operating the input unit.
- an input unit such as a keyboard 108, a mouse 109, and a touch panel (not shown). That is, the input interface 104 receives an external input such as an operation command given by the user operating the input unit.
- the display controller 105 is connected to a display 110, which is a typical example of a display unit, and controls display on the display 110. That is, the display controller 105 displays the result of image processing by the CPU 101 to the user.
- the display 110 is, for example, an LCD (Liquid Crystal Display) or a CRT (Cathode Ray Tube).
- the data reader / writer 106 mediates data transmission between the CPU 101 and the recording medium 111. That is, the recording medium 111 is distributed in a state where a program executed by the related information output device 201 is stored, and the data reader / writer 106 reads the program from the recording medium 111. Further, the data reader / writer 106 writes the processing result in the related information output device 201 into the recording medium 111 in response to the internal command of the CPU 101.
- the recording medium 111 may be, for example, a general-purpose semiconductor storage device such as CF (Compact Flash) and SD (Secure Digital), a magnetic storage medium such as a flexible disk, or a CD-ROM (Compact Disk Read Only). Memory).
- the communication interface 107 mediates data transmission between the CPU 101 and a personal computer, a server device or the like.
- the communication interface 107 typically has an Ethernet (registered trademark) or USB (Universal Serial Bus) communication function.
- Ethernet registered trademark
- USB Universal Serial Bus
- the related information output device 201 may be connected to another output device such as a printer as necessary.
- FIG. 2 is a block diagram showing a control structure provided by the related information output device according to the first embodiment of the present invention.
- Each block of the related information output device 201 shown in FIG. 2 is provided by developing a program (code) stored in the hard disk 103 in the main memory 102 and causing the CPU 101 to execute it.
- a program code
- some or all of the modules shown in FIG. 2 may be provided by firmware implemented in hardware.
- part or all of the control structure shown in FIG. 2 may be realized by dedicated hardware and / or a wiring circuit.
- the related information output device 201 includes a CPU 101 and a related information storage unit 56.
- the CPU 101 refers to the input unit 51 that inputs one or more related source information to be input, and the related information storage unit 56, and related information that searches related destination information and the like based on the related source information Based on a combination of two or more of the search source 52, the recognition source calculation unit 53 for calculating the related source information, the related destination information and the link recognition level, and the related source information, the related destination information and the link recognition level A high-interest information narrowing unit 54 that narrows down the information that is of interest to the user, and an output unit 55 that outputs the narrowed-down related information.
- the related information storage unit 56 corresponds to, for example, the main memory 102 or the hard disk 103 shown in FIG.
- the output unit 55 outputs related information to, for example, the hard disk 103 or the display 110 shown in FIG.
- the CPU 101 is generally composed of many units in addition to the components shown in FIG. 2, but units not related to the present invention are not shown for simplicity of explanation.
- the related information storage unit 56 holds related source information, related destination information, and their appearance frequency (hereinafter also simply referred to as frequency), that is, the number of links.
- frequency the appearance frequency
- hypertext documents are stored, and a certain hypertext document and its hyperlink destination document may be set as related source information and related destination information, respectively.
- a feature word of a certain hypertext document and a feature word of the hyperlink destination document may be used as related source information and related destination information, respectively.
- the input unit 51 receives one or more pieces of related source information and passes them to the related information search unit 52.
- the related information search unit 52 receives one or more pieces of related source information from the input unit 51, refers to the related information storage unit 56, and acquires related information including related source information that matches the received related source information. It passes to the recognition calculation unit 53.
- This related information includes related source information, related destination information, and their frequencies.
- the recognition level calculation unit 53 receives the related information that is the search result from the related information search unit 52, and calculates the related source information, the related destination information, and the link recognition level.
- the degree-of-recognition calculation unit 53 is related source information with respect to the overall frequency of the related information storage unit 56, the ratio of the frequency of each of the related destination information and the link, or related destination information with respect to the frequency of the entire search result, the related source The ratio of the frequency of each information and link is calculated as the degree of recognition.
- the high interest information narrowing unit 54 narrows down related information based on each recognition degree calculated by the recognition degree calculation unit 53.
- these related information is natural related information that everyone knows.
- any of the recognition levels is low, these pieces of information are considered to be information that the user does not know but does not want to know so much.
- the link (relevance) is low, the information is at least one of the related destination information and the related source information for the user. I know, but the relationship is thought to be surprising information.
- the highly interested information narrowing unit 54 determines that these pieces of information are unexpected when both the related source information and the related destination information have high recognition and the link recognition is low. Further, the high interest information narrowing section 54 determines that the information is unexpected when one of the related source information and the related destination information is high, the other is low, and the link is low. In other cases, it is determined that there is no unexpectedness. Further, the high interest information narrowing unit 54 may include the degree of unexpectedness in these rules, as will be described later.
- the output unit 55 receives related information from the highly interested information narrowing unit 54 and outputs the received related information.
- the related information output method according to the first embodiment of the present invention is implemented by operating the related information output device 201. Therefore, the description of the related information output method according to the first embodiment of the present invention is replaced with the following description of the operation of the related information output apparatus 201. In the following description, FIG. 2 is referred to as appropriate.
- FIG. 3 is a flowchart showing an operation procedure when the related information output apparatus according to the first embodiment of the present invention performs related information output processing.
- FIG. 4 is a diagram showing an example of stored information in the related information storage unit in the related information output device according to the first embodiment of the present invention.
- the related information storage unit 56 has tables T1 and T2.
- the table T1 stores related information, and one record indicates one related information.
- Each related information in the table T1 includes related source information, related destination information, and link frequency.
- the appearance frequency of actual information corresponding to the related source information and the related destination information is stored, and one record indicates one information.
- the related source information and the related destination information may be IDs of the respective information.
- hypertext documents are stored, and in the hypertext document, characteristic words included in the hyperlink source are used as related source information, and characteristic words included in the hyperlink destination hypertext are used as related destination information.
- the appearance frequency of each word may be the appearance frequency of the related source information and the related destination information.
- hypertext documents are accumulated, categories are associated with these hypertext documents, and there is a link from one category of hypertext document to another category, this link is used as related information.
- the number of hypertext documents in the category may be used as the appearance frequency. That is, the related information storage unit 56 only needs to store related source information, related destination information, data that can calculate the frequency thereof, and data that can calculate the information and the frequency thereof.
- the input unit 51 receives one or more pieces of related source information. For example, “Nippon Budokan” is input (step S1).
- the related information search unit 52 searches for related information (step S2). More specifically, the related information search unit 52 refers to the related information storage unit 56, acquires a record containing the related source information passed from the input unit 51, and returns it as a search result.
- the related information search unit 52 has 50 links to related information (“Nippon Budokan” and “Momoe Yamaguchi”) and related information (“Nippon Budokan”).
- the degree-of-recognition calculation unit 53 calculates the degree of recognition of each of the related source information, the related destination information, and the link for each related information (step S3).
- the degree of recognition increases as the frequency of each of the related source information, the related destination information, and the link increases.
- the degree-of-recognition calculation unit 53 calculates the relative frequency with respect to the total frequency of all the related information accumulated in the related information storage unit 56, and the frequency sum of all the related information of the search result by the related information search unit 52.
- the relative frequency, the relative difference from the average appearance frequency of all the related information stored in the related information storage unit 56, or the appearance probability in the estimated distribution of the entire related information is calculated as the degree of recognition.
- the degree-of-recognition calculation unit 53 uses each category of related source information and related destination information, calculates the appearance probability within the category as the degree of recognition of each of the related source information and related destination information, Appearance probability is calculated as link recognition (step S3).
- the degree-of-recognition calculation unit 53 calculates the relative difference from the average appearance frequency according to the following formulas, for example, the degree of recognition of the related source information A, the degree of recognition of the related destination information B, and their links (A, Calculate as B) awareness.
- Awareness (related source information A) frequency (related source information A) ⁇ average appearance frequency (related source information)
- Awareness (related information B) frequency (related information B)-average appearance frequency (related information)
- Awareness (link (A, B)) frequency (link (A, B))-average appearance frequency (link)
- the degree-of-recognition calculation unit 53 determines that the degree of recognition of the information is high and major if the calculated degree of recognition is equal to or greater than a predetermined threshold value set in advance.
- the recognition degree calculation unit 53 may calculate the average appearance frequency in advance and store it.
- the average appearance frequency of each of the related source information and the related destination information is 150, and the average appearance frequency of the link is 20.
- the related source information is a random variable
- the total frequency of the related source information is n
- the probability that the specific related source information appears is p
- the frequency of the specific related source information is k.
- the total frequency n of related source information is the sum of the frequencies of all related source information stored in the related information storage unit 56.
- the sum of 5000, 30, 500, and so on which is the frequency of “Nippon Budokan”, “Fukudaya Ryokan”, “Mori no Musicado”, etc. It becomes frequency.
- the appearance frequency of specific related source information A is nA.
- This probability value is used as the degree of recognition of related source information.
- link recognition is calculated as follows.
- the link is considered as a random variable
- the total frequency of the link is n
- the probability that the link appears is p
- the frequency of a specific link is kAB.
- the total link frequency n is the sum of the frequencies of all links stored in the related information storage unit 56.
- the related information accumulating unit 56 shown in FIG. 4 (“Nippon Budokan”, “Momoe Yamaguchi”), (“Nippon Budokan”, “Self Defense Force”), (“Nippon Budokan”, “Hikari Matsunaga”),.
- the sum of the frequencies 50, 2, 20,... Is the total frequency of the link.
- the appearance frequency of the specific link (A, B) is nAB.
- the total frequency of links is calculated by the total frequency of related source information ⁇ the total frequency of related destination information.
- the degree-of-recognition calculation unit 53 determines that the degree of recognition of the information is high and major when the degree of recognition obtained by performing the above calculation is equal to or greater than a predetermined threshold.
- the related information output device can also calculate link recognition using category information so that link recognition can be appropriately calculated. is there. That is, the link is considered for each category, and the recognition degree of the link between categories is calculated.
- the distribution is calculated for the category of related source information and the category of related destination information in the same manner as the related source information method described above.
- the degree of recognition between the category CA of the related source information A and the category CB of the related destination information B forming the link (A, B) is calculated.
- the related source information A or the related source information B belongs to a plurality of categories, an average value, a maximum value, or a minimum value of the degree of recognition of the links between the categories is obtained.
- nCAB is the frequency at which a link exists between category CA and category CB.
- n is the total frequency of link between category CA and category CB, that is, (number of related source information belonging to category CA) ⁇ (number of related destination information belonging to category CB).
- the high interest information screening unit 54 receives the recognition level from the recognition level calculation unit 53, and by referring to the rules incorporated in the high interest information screening unit 54 itself, is the related information surprising? It is determined whether or not (step S4).
- FIG. 5 is a diagram illustrating an example of the rules for determining the unexpectedness in the highly interested information narrowing unit 54.
- this rule indicates whether or not there is an unexpectedness for each combination of whether the related source information, the related destination information, and the link are major.
- the high interest information narrowing unit 54 determines that the combination of the related source information and the related destination information, that is, the related information is Judged as unexpected.
- the related information (“Nippon Budokan”, “Momoe Yamaguchi”) is all related source information, related destination information, and links are all major, and everyone knows. It is determined that there is no surprise.
- related information (“Nippon Budokan”, “SDF”) is related source information is major, related destination information is also major, and links are minor, so each information is known, but related information It is surprising that there is. Therefore, the highly interested information narrowing unit 54 determines that the related information is highly unexpected and selects it as information to be output.
- related information (“Nippon Budokan”, “Matsunaga Hikaru”) is related source information is major, related destination information is minor, and link is minor. Therefore, the highly interested information narrowing unit 54 determines that the unexpectedness of the related information is moderate, and selects the information to be output.
- the high interest information narrowing section 54 shows that the unexpectedness of the related information (“Nippon Budokan”, “Self Defense Force”) is high, and the unexpectedness of the related information (“Nippon Budokan”, “Matsumoto Hikaru”) Information indicating the presence is passed to the output unit 55 (step S4).
- the output unit 55 receives information from the high interest narrowing unit and outputs it (step S5).
- the output unit 55 displays the related information (“Nippon Budokan”, “Self Defense Force”) and related information (“Nippon Budokan”, “Hikari Matsumoto”) on the display 110.
- Non-Patent Documents 1 and 2 have a problem in that it is impossible to present unexpected information that attracts the user's interest.
- the related information search unit 52 receives the related source information from the input unit 51, and a plurality of related destinations related to the related source information
- the information, the appearance frequency of the related source information, the appearance frequency of each related destination information, and the appearance frequency of the relation between the related source information and each related destination information are acquired from the related information storage unit 56.
- the degree-of-recognition calculation unit 53 calculates the degree of recognition of related source information, the degree of recognition of each acquired related destination information, and the degree of recognition of each relationship based on each acquired appearance frequency.
- the high interest information narrowing unit 54 selects one or more related destination information from among the related destination information based on the recognition level of the related source information, the recognition level of the related destination information, and the recognition level of the relationship, and outputs Output to unit 55.
- the highly interested information narrowing unit 54 recognizes the related source information from the related destination information and the related information. At least one of the recognition degrees of the prior information is higher than a predetermined threshold, that is, major, and the related destination information whose relationship recognition is lower than the predetermined threshold, that is, minor is selected as the unexpected information.
- Such a configuration makes it possible to appropriately narrow down unexpectedly high interest information from a plurality of related information.
- the highly interested information narrowing unit 54 further recognizes related source information from the selected related destination information as shown in FIG.
- the related destination information whose degree of recognition of the degree and the related destination information is higher than a predetermined threshold and whose related recognition is lower than the predetermined threshold is selected as related destination information that is highly unexpected.
- Such a configuration makes it possible to present more unexpectedly high interest information to the user and further improve the user's satisfaction.
- the recognition level calculation unit 53 is a target for calculating the related source information, each acquired related destination information, and the recognition level among the respective relationships.
- the recognition level calculation unit 53 For the information, the sum of the appearance frequencies of all the related source information stored in the related information storage unit 56, the sum of the appearance frequencies of all the related destination information stored in the related information storage unit 56, and the related information storage unit 56
- the total appearance frequency corresponding to the target information is acquired from the total appearance frequencies of all the accumulated relationships.
- the degree-of-recognition calculation unit 53 acquires all the related source information stored in the related information storage unit 56, all the related destination information, and the appearance frequency of all the relationships via the related information search unit 52, From the acquired appearance frequencies, the total appearance frequency of the related source information, the total appearance frequency of the related destination information, and the total appearance frequency of the link are calculated. Then, the degree-of-recognition calculation unit 53 estimates the appearance probability distribution from the acquired appearance frequency sum, and uses the estimated appearance probability distribution to calculate the appearance probability of the target information as the degree of recognition of the target information.
- the degree-of-recognition calculation unit 53 calculates the average appearance frequency from the total appearance frequency acquired as described above, and the calculated average appearance frequency and The difference from the appearance frequency of the target information is calculated as the degree of recognition of the target information.
- the related information search unit 52 further stores the category information of the related source information and the category information of each related destination information related to the related information storage unit 56. Get from. Then, the degree-of-recognition calculation unit 53 estimates the appearance probability distribution from the sum of the appearance frequencies of the relations between the categories of the relation source information and the relation destination information corresponding to the target relation that is the relation for which the degree of recognition is to be calculated. The appearance probability of the relationship between categories is calculated as the recognition degree of the target relationship using the appearance probability distribution.
- the link recognition level can be appropriately calculated from the viewpoint of the category, so that high interest information can be narrowed down with higher accuracy.
- the recognition degree calculation unit 53 when at least one of the related source information and the related destination information corresponding to the target relationship belongs to a plurality of categories
- the average value, the maximum value, or the minimum value of the degree of recognition calculated for each category is set as the degree of recognition of the target relationship.
- the high interest information narrowing unit 54 selects one or a plurality of related information from among the related destination information based on the recognition level of the related source information, the recognition level of the related destination information, and the recognition level of the relationship.
- the high interest information screening unit 54 selects one or more related destinations from among the related destination information based on a combination of two or more of the recognition level of the related source information, the recognition level of the related destination information, and the recognition level of the relationship.
- the structure which selects information may be sufficient.
- the highly interested information narrowing unit 54 selects any two of the related source information, the related destination information, and the relationship, for example, the degree of recognition of one of the selected is higher than a predetermined threshold, and the other The related destination information whose degree of recognition is lower than a predetermined threshold is selected from each related destination information.
- the high interest information narrowing unit 54 selects related source information and related destination information, and determines that the related information is unexpected when the related source information is major and the related destination information is minor. To do. Further, the highly interested information narrowing unit 54 selects the related source information and the link, and when the related source information is major and the link is minor, the related information is determined to be unexpected. Further, the high interest information screening unit 54 selects the related destination information and the link, and determines that the related information is unexpected when the related destination information is major and the link is minor.
- the related information output device is configured to include the related information storage unit 56, the present invention is not limited to this.
- the related information storage unit 56 may be provided outside the related information output device 201.
- the output unit 55 is configured to output related source information and related destination information as related information.
- the present invention is not limited to this. Absent.
- the output unit 55 may be configured to output at least the related destination information, and may be configured to further output other information in addition to the related source information and the related destination information.
- the recognition degree calculation unit 53 includes, for example, all related source information, all related destination information, and all stored in the related information storage unit 56. Is obtained via the related information search unit 52, and the total appearance frequency of related source information, the total appearance frequency of related destination information, and the total appearance frequency of links are calculated from the acquired appearance frequencies.
- the related information storage unit 56 stores the total appearance frequency of the related source information, the total appearance frequency of the related destination information, and the total appearance frequency of the link
- the recognition degree calculation unit 53 stores the related information via the related information search unit 52. The configuration may be such that these appearance frequency sums are acquired from the unit 56.
- the threshold value for the recognition degree calculation unit 53 to determine the major or minor of the related information may be common to the related source information, the related destination information, and the link, or may be a different value.
- the recognition degree calculation unit 53 is configured to calculate link recognition degree using category information, but the present invention is not limited to this. is not.
- the degree-of-recognition calculation unit 53 may further calculate the degree of recognition of the category of the related source information, or may further calculate the degree of recognition of the category of the related destination information.
- the related information search unit 52 acquires the category information of the related source information and the category information of each related destination information from the related information storage unit 56. Then, the degree-of-recognition calculation unit 53 sets the category information of all the related-source information stored in the related-information storage unit 56 for the target information for which the degree-of-recognition is to be calculated. The sum of the appearance frequencies and the sum of the appearance frequencies corresponding to the target information among the sum of the appearance frequencies of the categories of all the related destination information accumulated in the related information accumulation unit 56 are acquired. The degree-of-recognition calculation unit 53 estimates the appearance probability distribution from the acquired appearance frequency sum, and calculates the appearance probability of the category of the target information as the degree of recognition of the target information using the estimated appearance probability distribution. Alternatively, the recognition degree calculation unit 53 calculates an average appearance frequency from the acquired appearance frequency sum, and calculates a difference between the calculated average appearance frequency and the appearance frequency of the category of the target information as the recognition degree of the target information.
- the recognition level of the related source information and the recognition level of the related destination information can be appropriately calculated from the viewpoint of the category, so that the high interest information can be narrowed down with higher accuracy.
- the recognition degree calculation unit 53 sets the average value, maximum value, or minimum value of the recognition degree calculated for each category as the recognition degree of the target information. . With such a configuration, even when there are a plurality of categories, the degree of recognition can be calculated appropriately.
- the present embodiment relates to a related information output device in which related content is added as a criterion for determination of unexpectedness compared to the related information output device according to the first embodiment.
- the contents other than those described below are the same as those of the related information output apparatus according to the first embodiment.
- the related information output device In the related information output device according to the first embodiment of the present invention, attention is paid to related source information, related destination information, and links, but in addition to these, related content is also important. For example, consider a geographical location and a person associated with that location. It is not surprising that the related source information “Mori no Musicado” and the related destination information “Seiji Ozawa” are output as related information. This is because the related source information is a concert hall, the related destination information is a conductor, and therefore the existence of the relationship is known because it can be assumed that a concert is performed. However, if the related content is “Mori no Kondo was designed by Seiji Ozawa”, this is surprising and interesting information. This is because the related content is different from the known inferred information.
- Non-Patent Documents 1 and 2 do not take into account the user's interest because they do not consider the major and minor degrees of the combination of two or more of the related source information, related destination information, link, and related content. There was a problem that it was impossible to present unexpected information.
- FIG. 6 is a block diagram showing a control structure provided by the related information output device according to the second embodiment of the present invention.
- Each block of the related information output device 202 shown in FIG. 6 is provided by developing a program (code) stored in the hard disk 103 in the main memory 102 and causing the CPU 101 to execute it.
- a program code
- some or all of the modules shown in FIG. 6 may be provided by firmware implemented in hardware.
- part or all of the control structure shown in FIG. 6 may be realized by dedicated hardware and / or a wiring circuit.
- the related information output device 202 includes a related information storage unit 66 instead of the related information storage unit 56, as compared to the related information output device according to the first embodiment of the present invention.
- the CPU 101 has, as its control structure, a recognition level calculation unit 63 and a high interest level instead of the recognition level calculation unit 53 and the high interest level information narrowing unit 54.
- An information screening unit 64 is provided.
- the related information storage unit 66 corresponds to, for example, the main memory 102 or the hard disk 103 shown in FIG.
- the CPU 101 is generally composed of many units in addition to the components shown in FIG. 6, but units not related to the present invention are not shown in order to simplify the description.
- FIG. 7 is a diagram showing an example of stored information in the related information storage unit in the related information output device according to the second embodiment of the present invention.
- the related information storage unit 66 has a table T3 instead of the table T1 as compared with the related information output device according to the first embodiment of the present invention.
- the table T3 stores related information, and one record indicates one related information.
- the table T3 is different from the table T1 in that it includes not only related source information, related destination information, and frequency but also related contents as related information.
- the first record shows that the related content of “Nippon Budokan” and “Momoe Yamaguchi” is “concert”.
- the second record indicates that the related source information and the related destination information are the same as the first record, but the related content is “location”.
- the recognition level calculation unit 63 is different from the recognition level calculation unit 53 in that the recognition level of related content is further calculated.
- the degree-of-recognition calculation unit 63 measures the degree of recognition of the related content depending on whether or not the related content frequently appears in the related source information and the related destination information, or in these categories.
- the category of related source information is “Hall” and the category of related destination information is “Entertainer”.
- the frequency of the related content “concert” is as high as 47, the degree of recognition is high.
- the frequency of the related content “design” is as low as 25, so the degree of recognition is low.
- the related information output device 202 stores related content that is considered to have high recognition level as related content in advance and the related content to be recognized includes the related content, it is determined that the recognition level is high. May be.
- the high interest information narrowing unit 64 is different from the high interest information narrowing unit 54 in that a rule that further considers the degree of recognition of related contents is applied.
- the related information output method according to the second embodiment of the present invention is implemented by operating the related information output device 202. Therefore, the description of the related information output method according to the second embodiment of the present invention is replaced with the following description of the operation of the related information output device 202. In the following description, FIG. 6 will be referred to as appropriate.
- FIG. 8 is a flowchart showing an operation procedure when the related information output apparatus according to the second embodiment of the present invention performs related information output processing.
- the input unit 51 receives one or more pieces of related source information. For example, “Nippon Budokan” is input (step S11).
- the related information search unit 52 searches for related information. However, it differs from the first embodiment of the present invention in that the search result includes related contents. That is, when the content stored in the second related information storage unit 206 is the example shown in FIG. 7, the related information search unit 52 determines that the link frequency of the related information (“Nippon Budokan”, “Momoe Yamaguchi”, “Concert”) is 47. The link frequency of related information (“Nippon Budokan”, “Momoe Yamaguchi”, “Location”) is returned as a search result (step S12).
- the degree-of-recognition calculation unit 63 calculates the degree of recognition of each of related source information, related source information, related destination information, links, and related contents.
- the relation source information, the relation destination information, and the link recognition degree are calculated in the same manner as in the first embodiment of the present invention.
- the degree-of-recognition calculation unit 63 relates the relative frequency with the entire related information, the relative frequency with the search result by the related information search unit 52, the relative difference with the average appearance frequency, or the appearance probability in the estimated distribution of the entire related information. Calculate as content awareness.
- the degree-of-recognition calculation unit 63 calculates the degree of recognition of the related content based on whether or not a preset keyword exists.
- the recognition level calculation unit 63 uses the category of the related source information and the related destination information, and calculates the appearance probability of the related content in the category of the related source information and the category of the related destination information as the recognition level of the related content (step S13).
- the degree of recognition is obtained from the appearance probability of the related content in the category of the related source information and the category of the related destination information according to the following formulas.
- Awareness (related content AB) frequency of related content AB of category CA of related source information A and category CB of related destination information B / link frequency of category CA of related source information A and category CB of related destination information B
- the degree-of-recognition calculation unit 63 determines that the degree of recognition of the information is high and major if the calculated degree of recognition is equal to or greater than a predetermined threshold value set in advance.
- the degree of recognition of related content in related information (“Nippon Budokan”, “Momoe Yamaguchi”, “Location”) is calculated as follows.
- Recognition (location) frequency of category “hall” and related content “location” “location” / link frequency of category “hall” and category “celebrity”
- the high interest information screening unit 64 receives the recognition level from the recognition level calculation unit 63 and refers to the rules incorporated in the high interest information screening unit 64 to determine whether or not the related information is unexpected. Is determined (step S14).
- FIG. 9 is a diagram illustrating an example of a rule for unexpectedness determination in the high interest information narrowing unit 64.
- this rule indicates whether or not there is an unexpectedness for each combination of whether the related source information, the related destination information, the link, and the related content are major.
- the highly interested information narrowing-down unit 64 determines that these pieces of information are unexpected when the related source information, the related destination information, and the link are highly recognized and the related content is low.
- the high interest information narrowing unit 64 determines that there is an unexpectedness when all of the related source information, the related destination information, and the links are low and the related content is high.
- the high interest information screening unit 64 determines that the information is unexpected when one of the related source information and the related destination information is high, the other is low, and the link is low. To do. Further, the highly interested information narrowing unit 64 determines that the information is unexpected when the recognition level of the related source information and the related destination information is high and the recognition level of at least one of the link and the related content is low. Then, the highly interested information narrowing-down unit 64 determines that the information is not unexpected in other cases.
- the high interest information narrowing section 64 may include the degree of unexpectedness in these rules.
- the high interest information narrowing unit 64 has a high degree of recognition of the related source information and the related destination information, and when the recognition level of at least one of the link and the related content is low, these are highly surprising information. to decide.
- the high interest information narrowing unit 64 has a high level of recognition of one of the related source information and the related destination information, and the other is low, the link is low, and the related content is high. Judge that the information is moderately surprising. With such a configuration, only relevant information with a high degree of unexpectedness can be passed to the output unit 55.
- the output unit 55 receives information from the high interest information narrowing unit 64 and outputs it (step S15).
- the related information search unit 52 receives the related source information from the input unit 51, and a plurality of related destinations related to the related source information Information, related source information and related contents of each related destination information, appearance frequency of related source information, appearance frequency of each related destination information, appearance frequency of relation between related source information and each related destination information, and each association
- the content appearance frequency is acquired from the related information storage unit 66.
- the degree-of-recognition calculation unit 63 determines the degree of recognition of related source information, the degree of recognition of each piece of related destination information acquired, the degree of recognition of each relation, and the degree of recognition of each piece of related content acquired based on each acquired appearance frequency.
- the high interest information narrowing unit 64 selects one or more related destinations from among the related destination information based on the recognition level of the related source information, the recognition level of the related destination information, the recognition level of the relationship, and the recognition level of the related content. Information is selected and output to the output unit 55.
- the high interest information narrowing unit 64 is selected such that the degree of recognition of the prior information and the degree of recognition of the relationship are higher than a predetermined threshold, and the degree of recognition of the related content is lower than the predetermined threshold.
- the high interest information narrowing unit 64 has a recognition level of related source information, a recognition level of related destination information, and a recognition level of relations that are lower than a predetermined threshold, and the recognition level of related content is lower than a predetermined threshold.
- Related destination information that is higher than a predetermined threshold is selected.
- the high interest information narrowing unit 64 has a recognition degree of the related source information and a recognition degree of the related destination information higher than a predetermined threshold among the related destination information, and the recognition degree of the relation and the recognition degree of the related content.
- the related destination information in which at least one is lower than a predetermined threshold is selected.
- the high interest information narrowing unit 64 has a recognition degree of the related source information and a recognition degree of the related destination information higher than a predetermined threshold value, and the other recognition degree is lower than the predetermined threshold value.
- related destination information whose relationship recognition level is lower than a predetermined threshold is selected.
- Such a configuration makes it possible to appropriately narrow down unexpectedly high interest information from a plurality of related information.
- the high-interest information narrowing-down unit 64 as shown in FIG. 9.
- the related destination information whose recognition degree of the related destination information is higher than a predetermined threshold value and at least one of the recognition degree of the relationship and the related content is lower than the predetermined threshold value is set to a high level in the selected related destination information. Select as related information that is highly unexpected.
- the high interest information narrowing unit 64 has one of the related source information and the related destination information that is higher than a predetermined threshold among the selected related destination information, and the other is higher than the predetermined threshold. Relevant information that is low, the degree of recognition of the relationship is lower than a predetermined threshold value, and the degree of recognition of the related content is higher than the predetermined threshold value is related as moderately highly relevant related information in the selected related destination information select.
- Such a configuration makes it possible to present unexpectedly high interest information to the user in stages, and to further improve the user's satisfaction.
- the related information search unit 52 further stores the category information of the related source information and the category information of each related destination information related to the related information storage unit 66. Get from. Then, the degree-of-recognition calculation unit 63 calculates the sum of the frequency of appearance of the relationship between the categories and the appearance of the related content for the relationship between the categories of the related source information and the related destination information corresponding to the related content for which the level of recognition should be calculated. The appearance probability of the related content is calculated based on the frequency, and the calculated appearance probability is set as the degree of recognition of the related content.
- the link recognition level can be appropriately calculated from the viewpoint of the category, so that high interest information can be narrowed down with higher accuracy.
- the high interest information narrowing unit 64 is configured to recognize the related source information, the related destination information, the related information, and the related content.
- the configuration is such that one or a plurality of related destination information is selected from each related destination information based on the degree, the present invention is not limited to this.
- the high interest information screening unit 64 selects from among each related destination information based on a combination of two or more of the recognition level of the related source information, the recognition level of the related destination information, the recognition level of the relationship, and the recognition level of the related content. It may be configured to select one or a plurality of related destination information.
- the narrowing-down rule in the first embodiment of the present invention can be applied to the related information output device according to the second embodiment of the present invention.
- the high interest information narrowing unit 64 selects the related source information and the related content, and when the related source information is major and the related content is minor, the related information is unexpected. Is determined.
- the high interest information screening unit 64 selects the related destination information and the related content, and determines that the related information is unexpected when the related destination information is major and the related content is minor.
- the high interest information narrowing unit 64 selects a link and related content, and determines that the related information is unexpected when the link is major and the related content is minor.
- the high interest information narrowing unit 64 is configured to determine the unexpectedness according to the rules shown in FIG. 9, but the present invention is not limited to this. is not.
- the high interest information filtering unit 64 for example, when the related source information, the related destination information, and the link are major and the related content is minor, and the related source information, the related destination information, and the link are minor, and the related content is It may be configured to determine that there is an unexpectedness only when it is a major.
- the present embodiment relates to a position information display device to which the related information output device according to the second embodiment is applied.
- the contents other than those described below are the same as those of the related information output apparatus according to the first embodiment or the second embodiment.
- FIG. 10 is a block diagram showing a control structure provided by the position information display device according to the third embodiment of the present invention.
- Each block of the position information display device 401 shown in FIG. 10 is provided by developing a program (code) stored in the hard disk 103 in the main memory 102 and causing the CPU 101 to execute it.
- a program code
- some or all of the modules shown in FIG. 10 may be provided by firmware implemented in hardware.
- part or all of the control structure shown in FIG. 10 may be realized by dedicated hardware and / or a wiring circuit.
- the position information display device 401 includes a CPU 101, a position information storage unit 305, and a related information storage unit 306.
- the CPU 101 includes an input unit 301, a position information search unit 302, a related information narrowing unit 303, and an output unit 304 as its control structure.
- the position information storage unit 305 and the related information storage unit 306 correspond to, for example, the main memory 102 or the hard disk 103 shown in FIG.
- the CPU 101 is generally composed of many units in addition to the components shown in FIG. 10, but units not related to the present invention are not shown in order to simplify the description.
- the location information storage unit 305 stores spot information such as stations, buildings, and temples together with the name, information specifying a location such as an address and latitude and longitude, and a category indicating classification.
- the related information storage unit 306 stores related information related to the position information. That is, the related information storage unit 306 uses the related source information as the position information, the related destination information as the name of an object related to the position (hereinafter also referred to as a related object), and the related content as the relationship between the position information and the related object. Is stored as keywords and sentences indicating.
- the location information search unit 302 refers to the location information storage unit 305 and searches for location information on the condition of the location information name, location information address, location information latitude and longitude, and location information category.
- the related information screening unit 303 corresponds to the related information search unit 52, the recognition degree calculation unit 63, and the high interest information screening unit 64 in the related information output device according to the second embodiment of the present invention.
- the related information narrowing-down unit 303 also filters the output of the high interest information narrowing-down unit 64 by using a category of related information and a keyword of related content.
- FIG. 11 is a diagram showing an example of input contents provided by the input unit in the position information display device according to the third embodiment of the present invention.
- the input unit 301 receives an input from the user. That is, the input unit 301 receives a search condition as information from the user.
- the input unit 301 includes a spatial refinement condition such as a position information name, address, and latitude / longitude, a position condition related to a position information category, a related object name, a related object category, and a related content keyword.
- a spatial refinement condition such as a position information name, address, and latitude / longitude
- a position condition related to a position information category e.g., a position condition related to a position information category
- a related object name e.g., a related object category
- a related content keyword e.g., a search conditions among conditions related objects are accepted.
- the input unit 301 receives a latitude / longitude, a range of position information, an address, a category of related objects, and a keyword of related content.
- the location information search unit 302 refers to the location information storage unit 305 and narrows down the location information.
- FIG. 12 is a diagram illustrating an example of accumulated information in the position information accumulation unit in the position information display device according to the third embodiment of the present invention.
- the location information storage unit 305 is a database that stores location information names, location information addresses, location information latitude and longitude, and location information categories.
- the location information search unit 302 searches for location information based on conditions related to location information. For example, it is assumed that “Nippon Budokan” and “Mori no Music Hall” were obtained as a result of the search.
- the related information narrowing-down unit 303 refers to the related information storage unit 306 by using the name of the position information and the condition relating to the related object as input, and narrows down to unexpected related information. For example, with “Nihon Budokan” as input, related information (“Nippon Budokan”, “Momoe Yamaguchi”, “Location”), related information (“Nippon Budokan”, “Self Defense Force”, “Music Association”), and related information ( “Forest of Music", “Seiji Ozawa”, “Design”).
- the output unit 304 displays the result that matches the search condition from the user, that is, the position information and the related information obtained by the position information search unit 302 and the related information narrowing unit 303.
- the output unit 304 outputs related information to, for example, the hard disk 103 or the display 110 shown in FIG.
- FIG. 13 is a diagram showing an example of output contents of the position information display device according to the third embodiment of the present invention.
- the output unit 304 outputs the position information list to the upper part (position information display area) of the screen of the display 110, and outputs the related object list to the lower part (related object information display area).
- the output unit 304 narrows the related object list at the lower part of the screen to only those related to the selected position information. Further, when the user selects a related object, the output unit 304 displays a detailed page of the selected related object.
- a related information output device for outputting information related to related source information, Receiving the related source information, a plurality of related destination information related to the related source information, an appearance frequency of the related source information, an appearance frequency of each related destination information, the related source information, and the respective A related information search unit for acquiring the appearance frequency of the relationship of related destination information from a related information storage unit provided inside or outside the related information output device; Based on each of the acquired appearance frequencies, the degree of recognition of the related source information, the degree of recognition of each of the related destination information acquired, and a degree of recognition calculating unit for calculating the degree of recognition of each of the relations; Select one or a plurality of related destination information from among the related destination information based on a combination of two or more of the recognition level of the related source information, the recognition level of the related destination information, and the recognition level of the relationship And a high interest information screening unit for outputting at least the selected related destination information.
- the highly interested information narrowing-down unit selects any two of the related source information, the related destination information, and the relationship, and the degree of recognition of one of the selected is higher than a predetermined threshold, and the degree of recognition of the other
- the related information output device according to supplementary note 1, wherein the related destination information that becomes lower than a predetermined threshold is selected from the related destination information.
- the high interest information narrowing-down unit has at least one of the degree of recognition of the related source information and the degree of recognition of the related destination information higher than a predetermined threshold, and the degree of recognition of the relationship is predetermined.
- the related information output device according to supplementary note 1 or supplementary note 2, wherein the relevant destination information that is lower than the threshold value is selected.
- the highly interested information narrowing unit further has a recognition degree of the related source information and a recognition degree of the related destination information higher than a predetermined threshold, and a recognition degree of the relationship is predetermined from the selected related destination information.
- the related information output device according to supplementary note 2 or supplementary note 3, wherein the relevant destination information that is lower than the threshold value is selected as highly relevant destination information that is highly unexpected.
- the degree-of-recognition calculation unit includes all the relations accumulated in the relational information accumulation part for the target information for which the degree-of-recognition of the relation source information, the acquired relation destination information and the relations to be calculated.
- the target among the sum of the appearance frequencies of the original information, the sum of the appearance frequencies of all the related destination information accumulated in the related information storage unit, and the sum of the appearance frequencies of all the relationships stored in the related information storage unit
- the appearance frequency total corresponding to the information is acquired, the appearance probability distribution is estimated from the acquired appearance frequency total, and the appearance probability of the target information is calculated as the recognition degree of the target information using the estimated appearance probability distribution
- the related information output device according to any one of appendix 1 to appendix 4.
- the degree-of-recognition calculation unit includes all the relations accumulated in the relational information accumulation part for the target information for which the degree-of-recognition of the relation source information, the acquired relation destination information and the relations to be calculated.
- the target among the sum of the appearance frequencies of the original information, the sum of the appearance frequencies of all the related destination information accumulated in the related information storage unit, and the sum of the appearance frequencies of all the relationships stored in the related information storage unit
- the total appearance frequency corresponding to the information is acquired, the average appearance frequency is calculated from the acquired total appearance frequency, and the difference between the calculated average appearance frequency and the appearance frequency of the target information is calculated as the degree of recognition of the target information
- the related information output device according to any one of appendix 1 to appendix 4.
- the related information search unit further acquires the category information of the related source information and the category information of each related destination information related from the related information storage unit,
- the degree-of-recognition calculation unit estimates an appearance probability distribution from a sum of appearance frequencies of relations between categories of the relation source information and the relation destination information corresponding to the target relation that is the relation for which the degree of recognition is to be calculated.
- the related information output device according to any one of appendix 1 to appendix 6, wherein the occurrence probability of the relationship between the categories is calculated as the recognition degree of the target relationship using the estimated occurrence probability distribution.
- the related information search unit further acquires the category information of the related source information and the category information of each related destination information related from the related information storage unit
- the recognition level calculation unit includes categories of all the related source information stored in the related information storage unit for the target information for which the recognition level is to be calculated among the related source information and the acquired related destination information. And the appearance frequency sum corresponding to the target information among the sum of the appearance frequencies of all the related destination information categories accumulated in the related information accumulation unit, and the obtained appearance frequency sum
- the related information search unit further acquires the category information of the related source information and the category information of each related destination information related from the related information storage unit
- the recognition level calculation unit includes categories of all the related source information stored in the related information storage unit for the target information for which the recognition level is to be calculated among the related source information and the acquired related destination information. And the appearance frequency sum corresponding to the target information among the sum of the appearance frequencies of all the related destination information categories accumulated in the related information accumulation unit, and the obtained appearance frequency sum
- the related information output device according to any one of appendix 1 to appendix 8, wherein an average appearance frequency is calculated, and a difference between the calculated average appearance frequency and the appearance frequency of the category of the target information is calculated as a recognition degree of the target information.
- the recognition degree calculation unit sets the average value, maximum value, or minimum value of the recognition degree calculated for each category as the recognition degree of the target information.
- the related information output device according to appendix 9 or appendix 10.
- the related information search unit receives the related source information, a plurality of related destination information related to the related source information, the related content of the related source information and each of the related destination information, and the appearance of the related source information
- a frequency, an appearance frequency of each related destination information, an appearance frequency of the relation between the related source information and each related destination information, and an appearance frequency of each related content are acquired from the related information storage unit
- the recognition level calculation unit based on each acquired appearance frequency, the recognition level of the related source information, the recognition level of each of the related destination information acquired, the recognition level of each of the relations and each of the acquired Calculate awareness of related content
- the high interest information narrowing-down unit is configured to determine each of the relations based on a combination of two or more of the degree of recognition of the relation source information, the degree of recognition of the relation destination information, the degree of recognition of the relation, and the degree of recognition of the related content.
- the related information output device according to any one of appendix 1 to appendix 11, wherein one or a plurality of the related destination
- the high-interest information narrowing unit has a recognition degree of the related source information, a recognition degree of the related destination information, and a recognition degree of the relationship higher than a predetermined threshold, and the related content information Selecting the related destination information whose degree of recognition is lower than a predetermined threshold; and Among the related destination information, the recognition level of the related source information, the recognition level of the related destination information, and the recognition level of the relationship are lower than a predetermined threshold value, and the recognition level of the related content is higher than a predetermined threshold value.
- the related information output device according to attachment 12, wherein the related destination information is selected.
- the high interest information narrowing unit further includes a degree of recognition of the related source information and a degree of recognition of the related destination information higher than a predetermined threshold, and a degree of recognition of the relationship and the relation Selecting the related destination information in which at least one of the content recognition levels is lower than a predetermined threshold; and Among the related destination information, the degree of recognition of the related source information and the degree of one of the related destination information are higher than a predetermined threshold, and the other is lower than a predetermined threshold, and the degree of recognition of the relationship Select the related destination information that becomes lower than a predetermined threshold,
- the high interest information narrowing-down unit has a degree of recognition of the related source information and a degree of recognition of the related destination information higher than a predetermined threshold, and a degree of recognition of the relationship and the related content from the selected related destination information
- the related destination information for which at least one of the recognition degrees of the information is lower than a predetermined threshold is selected as related destination information that is highly unexpected in the selected related destination information, and the related source
- One of the degree and the related destination information is higher than a predetermined threshold, the other is lower than a predetermined threshold, the degree of recognition of the relationship is lower than a predetermined threshold, and the degree of recognition of the related content is predetermined.
- the related information output device wherein the related destination information that is higher than a threshold value of the selected related destination information is selected as moderately unexpected related destination information from among the selected related destination information.
- the related information search unit further acquires the category information of the related source information and the category information of each related destination information related from the related information storage unit, The degree-of-recognition calculation unit, for the relationship between the categories of the related source information and the related destination information corresponding to the related content for which the degree of recognition is to be calculated,
- the related information output device according to any one of appendix 12 to appendix 14, wherein the appearance probability of the related content is calculated based on the appearance frequency of the related content, and the calculated appearance probability is used as the degree of recognition of the related content.
- a related information output method in a related information output device for outputting information related to related source information Receiving the related source information, a plurality of related destination information related to the related source information, an appearance frequency of the related source information, an appearance frequency of each related destination information, the related source information, and the respective Obtaining the appearance frequency of the relationship of related destination information from a related information storage unit provided inside or outside the related information output device; Calculating the degree of recognition of the related source information, the degree of recognition of each acquired related destination information, and the degree of recognition of each of the relations based on the acquired appearance frequencies; Select one or a plurality of related destination information from among the related destination information based on a combination of two or more of the recognition level of the related source information, the recognition level of the related destination information, and the recognition level of the relationship And outputting at least the selected related destination information.
- any two of the related source information, the related destination information, and the relationship are selected, and the degree of recognition of one of the selected is higher than a predetermined threshold, and the other
- the related information output method according to supplementary note 16 wherein the related destination information whose recognition level is lower than a predetermined threshold is selected from the related destination information.
- At least one of the recognition level of the related source information and the recognition level of the related destination information is higher than a predetermined threshold among the related destination information and the recognition of the relationship 18.
- the related information output method according to appendix 16 or appendix 17, wherein the related destination information whose degree is lower than a predetermined threshold is selected.
- the recognition degree of the related source information and the recognition degree of the related destination information are higher than a predetermined threshold from the selected related destination information, and the recognition of the relation is further performed.
- the related source information, each of the related destination information acquired, and the target information for which the degree of recognition of the relations should be calculated are stored in the related information storage unit.
- the appearance frequency total corresponding to the target information is acquired, the appearance probability distribution is estimated from the acquired appearance frequency total, and the appearance probability of the target information is recognized using the estimated appearance probability distribution.
- the related information output method according to any one of supplementary note 16 to supplementary note 19, which is calculated as a degree.
- the related source information, each of the related destination information acquired, and the target information for which the degree of recognition of the relations should be calculated are stored in the related information storage unit.
- the appearance frequency sum corresponding to the target information is acquired, the average appearance frequency is calculated from the acquired appearance frequency total, and the difference between the calculated average appearance frequency and the appearance frequency of the target information is recognized.
- the related information output method according to any one of supplementary note 16 to supplementary note 19, which is calculated as a degree.
- the sum of the appearance frequencies of the categories of information, the sum of the appearance frequencies of the categories of all the related information stored in the related information storage unit, and the appearance frequencies of the categories of all the relationships stored in the related information storage unit The appearance frequency total corresponding to the target information is acquired from the total, the appearance probability distribution is estimated from the acquired appearance frequency total, and the appearance probability of the category of the target information is calculated using the estimated appearance probability distribution.
- the related information output method according to any one of supplementary note 16 to supplementary note 23, which is calculated as the degree of recognition of the target information.
- the sum of the appearance frequencies of the categories of information, the sum of the appearance frequencies of the categories of all the related information stored in the related information storage unit, and the appearance frequencies of the categories of all the relationships stored in the related information storage unit The total appearance frequency corresponding to the target information is acquired from the total, the average appearance frequency is calculated from the acquired total appearance frequency, and the difference between the calculated average appearance frequency and the appearance frequency of the category of the target information is calculated as the target.
- the related information output method according to any one of supplementary note 16 to supplementary note 23, which is calculated as information recognition.
- the degree of recognition of the related source information, the degree of recognition of the related destination information, and the degree of recognition of the relationship are higher than a predetermined threshold from the respective pieces of related destination information, and The related destination information whose related content recognition level is lower than a predetermined threshold is selected, and among the related destination information, the recognition level of the related source information, the recognition level of the related destination information, and the relationship 28.
- the degree of recognition of the related source information and the degree of recognition of the related destination information are higher than a predetermined threshold from the related destination information, and Selecting the related destination information in which at least one of the degree of recognition of the relationship and the degree of recognition of the related content is lower than a predetermined threshold as the related destination information that is highly unexpected in the selected related destination information;
- the degree of recognition of the related source information and the degree of one of the related destination information are higher than a predetermined threshold, and the other is lower than a predetermined threshold, and the degree of recognition of the relationship
- Select the related destination information that becomes lower than a predetermined threshold Among the selected related destination information, the degree of recognition of the related source information and the degree of recognition of the related destination information are higher than a predetermined threshold, and at least one of the degree of recognition of the relationship and the degree of recognition of the related content is predetermined.
- the related destination information that is lower than the threshold value is selected as related destination information that is highly unexpected in the selected related destination information, and one of the recognition degree of the related source information and the related destination information
- the related information output method according to appendix 28 wherein information is selected as related information that is moderately surprising among the selected related information.
- a computer-readable recording medium recording a related information output program used in a related information output device for outputting information related to related source information, Receiving the related source information, a plurality of related destination information related to the related source information, an appearance frequency of the related source information, an appearance frequency of each related destination information, the related source information, and the respective Obtaining the appearance frequency of the relationship of related destination information from a related information storage unit provided inside or outside the related information output device; Calculating the degree of recognition of the related source information, the degree of recognition of each acquired related destination information, and the degree of recognition of each of the relations based on the acquired appearance frequencies; Select one or a plurality of related destination information from among the related destination information based on a combination of two or more of the recognition level of the related source information, the recognition level of the related destination information, and the recognition level of the relationship And a computer-readable recording medium recording a related information output program for causing a computer to execute at least the step of outputting the selected related destination information.
- any two of the related source information, the related destination information, and the relationship are selected, and the degree of recognition of one of the selected is higher than a predetermined threshold, and the other 32.
- At least one of the recognition level of the related source information and the recognition level of the related destination information is higher than a predetermined threshold among the related destination information and the recognition of the relationship
- the related source information, each of the related destination information acquired, and the target information for which the degree of recognition of the relations should be calculated are stored in the related information storage unit.
- the appearance frequency total corresponding to the target information is acquired, the appearance probability distribution is estimated from the acquired appearance frequency total, and the appearance probability of the target information is recognized using the estimated appearance probability distribution.
- the computer-readable recording medium according to any one of appendix 31 to appendix 34, calculated as a degree.
- the related source information, each of the related destination information acquired, and the target information for which the degree of recognition of the relations should be calculated are stored in the related information storage unit.
- the appearance frequency total corresponding to the target information is acquired, the average appearance frequency is calculated from the acquired appearance frequency total, and the difference between the calculated average appearance frequency and the appearance frequency of the target information is recognized.
- the computer-readable recording medium according to any one of appendix 31 to appendix 34, calculated as a degree.
- the sum of the appearance frequencies of the categories of information, the sum of the appearance frequencies of the categories of all the related information stored in the related information storage unit, and the appearance frequencies of the categories of all the relationships stored in the related information storage unit The appearance frequency total corresponding to the target information is acquired from the total, the appearance probability distribution is estimated from the acquired appearance frequency total, and the appearance probability of the category of the target information is calculated using the estimated appearance probability distribution.
- the computer-readable recording medium according to any one of appendix 31 to appendix 38, which is calculated as the degree of recognition of the target information.
- the sum of the appearance frequencies of the categories of information, the sum of the appearance frequencies of the categories of all the related information stored in the related information storage unit, and the appearance frequencies of the categories of all the relationships stored in the related information storage unit The total appearance frequency corresponding to the target information is acquired from the total, the average appearance frequency is calculated from the acquired total appearance frequency, and the difference between the calculated average appearance frequency and the appearance frequency of the category of the target information is calculated as the target Calculate as information awareness,
- the computer-readable recording medium according to any one of appendix 31 to appendix 38.
- the degree of recognition of the related source information, the degree of recognition of the related destination information, and the degree of recognition of the relationship are higher than a predetermined threshold from the respective pieces of related destination information, and The related destination information whose related content recognition level is lower than a predetermined threshold is selected, and among the related destination information, the recognition level of the related source information, the recognition level of the related destination information, and the relationship 43.
- the degree of recognition of the related source information and the degree of recognition of the related destination information are higher than a predetermined threshold from the related destination information, and the degree of recognition of the relationship And the related destination information in which at least one of the degree of recognition of the related content is lower than a predetermined threshold is selected as related destination information that is highly unexpected in the selected related destination information.
- the degree of recognition of the related source information and the degree of one of the related destination information are higher than a predetermined threshold, and the other is lower than a predetermined threshold, and the degree of recognition of the relationship Select the related destination information that becomes lower than a predetermined threshold, Among the selected related destination information, the degree of recognition of the related source information and the degree of recognition of the related destination information are higher than a predetermined threshold, and at least one of the degree of recognition of the relationship and the degree of recognition of the related content is predetermined.
- the related destination information that is lower than the threshold value is selected as related destination information that is highly unexpected in the selected related destination information, and one of the recognition degree of the related source information and the related destination information
- the computer-readable recording medium according to appendix 43 wherein information is selected as related destination information that is moderately surprising among the selected related destination information.
- the present invention has industrial applicability because it can be used for a related information retrieval device for position information, a keyword input auxiliary device for text search, and the like.
Abstract
Description
[概要]
前述のように、ユーザにとって既知情報であるか未知情報であるかは、ユーザの興味を引く情報であるかどうかに対して非常に重要な要素である。したがって、関連情報が既知情報であるか未知情報であるかを判断する際、関連元情報に関する判断、関連先情報に関する判断、および関連の存在に関する判断の3つの判断を行なうことが重要となる。非特許文献2に記載の技術では、関連の存在に関する判断は行われているが、関連元情報に関する判断、および関連先情報に関する判断が行われてない。
図1は、本発明の第1の実施の形態に係る関連情報出力装置の概略構成図である。図1を参照して、関連情報出力装置201は、携帯情報端末、パーソナルコンピュータおよびサーバ等の情報処理装置であり、演算処理部であるCPU(Central Processing Unit)101と、メインメモリ102およびハードディスク103と、入力インターフェイス104と、表示コントローラ105と、データリーダ/ライタ106と、通信インターフェイス107とを備える。これらの各部は、バス121を介して互いにデータ通信可能に接続される。
次に、関連情報出力装置201における各種機能を提供するための制御構造について説明する。
次に、本発明の第1の実施の形態に係る関連情報出力装置の動作について図面を用いて説明する。本発明の第1の実施の形態では、関連情報出力装置201を動作させることによって、本発明の第1の実施の形態に係る関連情報出力方法が実施される。よって、本発明の第1の実施の形態に係る関連情報出力方法の説明は、以下の関連情報出力装置201の動作説明に代える。なお、以下の説明においては、適宜図2を参照する。
認知度(関連元情報A)=頻度(関連元情報A)-平均出現頻度(関連元情報)
認知度(関連先情報B)=頻度(関連先情報B)-平均出現頻度(関連先情報)
認知度(リンク(A、B))=頻度(リンク(A、B))-平均出現頻度(リンク)
れる。
認知度(日本武道館)=5000-150=4850
認知度(山口百恵)=10000-150=9850
認知度(リンク(日本武道館、山口百恵))=50-20=30
認知度(日本武道館)=5000-150=4850
認知度(自衛隊)=200-150=50
認知度(リンク(日本武道館、自衛隊))=2-20=-18
認知度(日本武道館)=5000-150=4850
認知度(松永光)=30-150=-120
認知度(リンク(日本武道館、松永光))=20-20=0
P(X)=B(n,p,k)=nCk×p^k×(1-p)^(1-k)
Σ_(k=0,nA)P(X)=Σ_(k=0,nA){nCk×p^k×(1-p)^(1-k)}
p=関連元情報の平均頻度/関連元情報の全頻度
Σ_(k=0,nAB)P(X)=Σ_(k=0,nAB){nCk×p^k×(1-p)^(1-k)}
また、パラメータpは以下のように計算される。
p=リンクの出現頻度/リンクの全頻度
ここで、一般的に、Webページにおけるリンクは1つである場合が多く、関連元情報と関連先情報との関係(リンク)の頻度は1である場合が多い。このような場合でもリンクの認知度を適切に計算できるように、本発明の第1の実施の形態に係る関連情報出力装置では、カテゴリ情報を用いてリンクの認知度を計算することも可能である。すなわち、リンクをカテゴリ毎で考え、カテゴリ間リンクの認知度を計算する。
Σ_(k=0,nCAB)P(X)=Σ_(k=0,nCAB){nCk×p^k×(1-p)^(1-k)}
p=カテゴリ間にリンクが存在する数/全カテゴリ間のリンクの全頻度
積部56が関連元情報の出現頻度総和、関連先情報の出現頻度総和およびリンクの出現頻度総和を保存しており、認知度計算部53が、関連情報検索部52経由で関連情報蓄積部56からこれらの出現頻度総和を取得する構成であってもよい。
本実施の形態は、第1の実施の形態に係る関連情報出力装置と比べて意外性判断の基準として関連内容を追加した関連情報出力装置に関する。以下で説明する内容以外は第1の実施の形態に係る関連情報出力装置と同様である。
本発明の第1の実施の形態に係る関連情報出力装置では、関連元情報、関連先情報、およびリンクに着目したが、これらに加えて、関連内容も重要である。たとえば、地理的な場所とその場所に関連する人物とを考える。関連元情報「森の音楽堂」と関連先情報「小澤征爾」が関連情報として出力されてもあまり意外ではない。なぜならば、この関連元情報はコンサートホールであり、この関連先情報は指揮者であり、したがって、コンサートをやることを推測できるため関連の存在は既知だからである。しかしながら、関連内容が「森の音楽堂は小澤征爾が設計した」であれば、これは意外性が高く興味を引く情報である。なぜならば、関連内容は既知の推測した情報と異なるからである。
図6は、本発明の第2の実施の形態に係る関連情報出力装置が提供する制御構造を示すブロック図である。
次に、本発明の第2の実施の形態に係る関連情報出力装置の動作について図面を用いて説明する。本発明の第2の実施の形態では、関連情報出力装置202を動作させることによって、本発明の第2の実施の形態に係る関連情報出力方法が実施される。よって、本発明の第2の実施の形態に係る関連情報出力方法の説明は、以下の関連情報出力装置202の動作説明に代える。なお、以下の説明においては、適宜図6を参照する。
認知度(関連内容AB)=関連元情報AのカテゴリCAと関連先情報BのカテゴリCBの関連内容ABの頻度/関連元情報AのカテゴリCAと関連先情報BのカテゴリCBのリンク頻度
認知度(ロケ)=カテゴリ「ホール」とカテゴリ「芸能人」の関連内容「ロケ」の頻度/カテゴリ「ホール」とカテゴリ「芸能人」のリンク頻度
認知度(ロケ)=カテゴリ「宿」とカテゴリ「芸能人」の関連内容「ロケ」の頻度/カテゴリ「宿」とカテゴリ「芸能人」のリンク頻度
図9を参照して、このルールは、関連元情報、関連先情報、リンクおよび関連内容がそれぞれメジャーであるか否かの組み合わせ毎に、意外性があるか否かを示すものである。
本実施の形態は、第2の実施の形態に係る関連情報出力装置を適用した位置情報表示装置に関する。以下で説明する内容以外は第1の実施の形態または第2の実施の形態に係る関連情報出力装置と同様である。
関連元情報に関連する情報を出力するための関連情報出力装置であって、
前記関連元情報を受けて、前記関連元情報に関連する複数の関連先情報と、前記関連元情報の出現頻度と、関連する各前記関連先情報の出現頻度と、前記関連元情報および前記各関連先情報の関係の出現頻度とを前記関連情報出力装置の内部または外部に設けられた関連情報蓄積部から取得するための関連情報検索部と、
取得された各前記出現頻度に基づいて、前記関連元情報の認知度、取得された前記各関連先情報の認知度、および各前記関係の認知度を計算するための認知度計算部と、
前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度のうち、2つ以上の組み合わせに基づいて前記各関連先情報の中から1または複数の前記関連先情報を選択し、選択した前記関連先情報を少なくとも出力するための高関心情報絞込み部とを備える、関連情報出力装置。
前記高関心情報絞込み部は、前記関連元情報、前記関連先情報および前記関係のうちのいずれか2つを選択し、選択したうちの一方の認知度が所定の閾値より高く、他方の認知度が所定の閾値より低くなる前記関連先情報を前記各関連先情報の中から選択する、付記1に記載の関連情報出力装置。
前記高関心情報絞込み部は、前記各関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度の少なくとも一方が所定の閾値より高く、かつ前記関係の認知度が所定の閾値より低くなる前記関連先情報を選択する、付記1または付記2に記載の関連情報出力装置。
前記高関心情報絞込み部は、さらに、選択した前記関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度が所定の閾値より高く、かつ前記関係の認知度が所定の閾値より低くなる前記関連先情報を、高程度に意外性の高い関連先情報として選択する、付記2または付記3に記載の関連情報出力装置。
前記認知度計算部は、前記関連元情報、取得された前記各関連先情報および前記各関係のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報の出現頻度の総和、前記関連情報蓄積部に蓄積されたすべての関連先情報の出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関係の出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から出現確率分布を推定し、推定した前記出現確率分布を用いて、前記対象情報の出現確率を前記対象情報の認知度として計算する、付記1から付記4のいずれかに記載の関連情報出力装置。
前記認知度計算部は、前記関連元情報、取得された前記各関連先情報および前記各関係のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報の出現頻度の総和、前記関連情報蓄積部に蓄積されたすべての関連先情報の出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関係の出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から平均出現頻度を計算し、計算した前記平均出現頻度と前記対象情報の出現頻度との差を前記対象情報の認知度として計算する、付記1から付記4のいずれかに記載の関連情報出力装置。
前記関連情報検索部は、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度計算部は、前記認知度を計算すべき前記関係である対象関係に対応する前記関連元情報および前記関連先情報の、カテゴリ間の関係の出現頻度の総和から出現確率分布を推定し、推定した前記出現確率分布を用いて、前記カテゴリ間の関係の出現確率を前記対象関係の認知度として計算する、付記1から付記6のいずれかに記載の関連情報出力装置。
前記認知度計算部は、前記対象関係に対応する前記関連元情報および前記関連先情報の少なくとも一方が複数のカテゴリに属している場合には、各前記カテゴリについて計算した前記認知度の平均値、最大値または最小値を前記対象関係の認知度とする、付記7に記載の関連情報出力装置。
前記関連情報検索部は、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度計算部は、前記関連元情報および取得された前記各関連先情報のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報のカテゴリの出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関連先情報のカテゴリの出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から出現確率分布を推定し、推定した前記出現確率分布を用いて、前記対象情報のカテゴリの出現確率を前記対象情報の認知度として計算する、付記1から付記8のいずれかに記載の関連情報出力装置。
前記関連情報検索部は、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度計算部は、前記関連元情報および取得された前記各関連先情報のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報のカテゴリの出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関連先情報のカテゴリの出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から平均出現頻度を計算し、計算した平均出現頻度と前記対象情報のカテゴリの出現頻度との差を前記対象情報の認知度として計算する、付記1から付記8のいずれかに記載の関連情報出力装置。
前記認知度計算部は、前記対象情報が複数のカテゴリに属している場合には、各前記カテゴリについて計算した前記認知度の平均値、最大値または最小値を前記対象情報の認知度とする、付記9または付記10に記載の関連情報出力装置。
前記関連情報検索部は、前記関連元情報を受けて、前記関連元情報に関連する複数の関連先情報と、前記関連元情報および各前記関連先情報の関連内容と、前記関連元情報の出現頻度と、前記各関連先情報の出現頻度と、前記関連元情報および前記各関連先情報の関係の出現頻度と、各前記関連内容の出現頻度とを前記関連情報蓄積部から取得し、
前記認知度計算部は、取得された各前記出現頻度に基づいて、前記関連元情報の認知度、取得された前記各関連先情報の認知度、各前記関係の認知度および取得された前記各関連内容の認知度を計算し、
前記高関心情報絞込み部は、前記関連元情報の認知度、前記関連先情報の認知度、前記関係の認知度および前記関連内容の認知度のうち、2つ以上の組み合わせに基づいて前記各関連先情報の中から1または複数の前記関連先情報を選択し、選択した前記関連先情報を少なくとも出力する、付記1から付記11のいずれかに記載の関連情報出力装置。
前記高関心情報絞込み部は、前記各関連先情報の中から、前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度が所定の閾値より高く、かつ前記関連内容の認知度が所定の閾値より低くなる前記関連先情報を選択し、かつ、
前記各関連先情報の中から、前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度が所定の閾値より低く、かつ前記関連内容の認知度が所定の閾値より高くなる前記関連先情報を選択する、付記12に記載の関連情報出力装置。
前記高関心情報絞込み部は、さらに、前記各関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度が所定の閾値より高く、かつ前記関係の認知度および前記関連内容の認知度の少なくとも一方が所定の閾値より低くなる前記関連先情報を選択し、かつ、
前記各関連先情報の中から、前記関連元情報の認知度および前記関連先情報の一方の認知度が所定の閾値より高く、他方の認知度が所定の閾値より低く、かつ前記関係の認知度が所定の閾値より低くなる前記関連先情報を選択し、
前記高関心情報絞込み部は、選択した前記関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度が所定の閾値より高く、かつ前記関係の認知度および前記関連内容の認知度の少なくとも一方が所定の閾値より低くなる前記関連先情報を、選択した前記関連先情報の中で高程度に意外性の高い関連先情報として選択し、かつ、前記関連元情報の認知度および前記関連先情報の一方の認知度が所定の閾値より高く、他方の認知度が所定の閾値より低く、前記関係の認知度が所定の閾値より低く、かつ前記関連内容の認知度が所定の閾値より高くなる前記関連先情報を、選択した前記関連先情報の中で中程度に意外性の高い関連先情報として選択する、付記13に記載の関連情報出力装置。
前記関連情報検索部は、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度計算部は、前記認知度を計算すべき前記関連内容に対応する前記関連元情報および前記関連先情報の各々のカテゴリ間の関係について、前記カテゴリ間の関係の出現頻度の総和および前記関連内容の出現頻度に基づいて前記関連内容の出現確率を計算し、計算した前記出現確率を前記関連内容の認知度とする、付記12から付記14のいずれかに記載の関連情報出力装置。
関連元情報に関連する情報を出力するための関連情報出力装置における関連情報出力方法であって、
前記関連元情報を受けて、前記関連元情報に関連する複数の関連先情報と、前記関連元情報の出現頻度と、関連する各前記関連先情報の出現頻度と、前記関連元情報および前記各関連先情報の関係の出現頻度とを前記関連情報出力装置の内部または外部に設けられた関連情報蓄積部から取得するステップと、
取得された各前記出現頻度に基づいて、前記関連元情報の認知度、取得された前記各関連先情報の認知度、および各前記関係の認知度を計算するステップと、
前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度のうち、2つ以上の組み合わせに基づいて前記各関連先情報の中から1または複数の前記関連先情報を選択し、選択した前記関連先情報を少なくとも出力するステップとを含む、関連情報出力方法。
前記関連先情報を選択するステップにおいては、前記関連元情報、前記関連先情報および前記関係のうちのいずれか2つを選択し、選択したうちの一方の認知度が所定の閾値より高く、他方の認知度が所定の閾値より低くなる前記関連先情報を前記各関連先情報の中から選択する、付記16に記載の関連情報出力方法。
前記関連先情報を選択するステップにおいては、前記各関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度の少なくとも一方が所定の閾値より高く、かつ前記関係の認知度が所定の閾値より低くなる前記関連先情報を選択する、付記16または付記17に記載の関連情報出力方法。
前記関連先情報を選択するステップにおいては、さらに、選択した前記関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度が所定の閾値より高く、かつ前記関係の認知度が所定の閾値より低くなる前記関連先情報を、高程度に意外性の高い関連先情報として選択する、付記17または付記18に記載の関連情報出力方法。
前記認知度を計算するステップにおいては、前記関連元情報、取得された前記各関連先情報および前記各関係のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報の出現頻度の総和、前記関連情報蓄積部に蓄積されたすべての関連先情報の出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関係の出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から出現確率分布を推定し、推定した前記出現確率分布を用いて、前記対象情報の出現確率を前記対象情報の認知度として計算する、付記16から付記19のいずれかに記載の関連情報出力方法。
前記認知度を計算するステップにおいては、前記関連元情報、取得された前記各関連先情報および前記各関係のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報の出現頻度の総和、前記関連情報蓄積部に蓄積されたすべての関連先情報の出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関係の出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から平均出現頻度を計算し、計算した前記平均出現頻度と前記対象情報の出現頻度との差を前記対象情報の認知度として計算する、付記16から付記19のいずれかに記載の関連情報出力方法。
前記各関連先情報および前記各出現頻度を取得するステップにおいては、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度を計算するステップにおいては、前記認知度を計算すべき前記関係である対象関係に対応する前記関連元情報および前記関連先情報の、カテゴリ間の関係の出現頻度の総和から出現確率分布を推定し、推定した前記出現確率分布を用いて、前記カテゴリ間の関係の出現確率を前記対象関係の認知度として計算する、付記16から付記21のいずれかに記載の関連情報出力方法。
前記認知度を計算するステップにおいては、前記対象関係に対応する前記関連元情報および前記関連先情報の少なくとも一方が複数のカテゴリに属している場合には、各前記カテゴリについて計算した前記認知度の平均値、最大値または最小値を前記対象関係の認知度とする、付記22に記載の関連情報出力方法。
前記各関連先情報および前記各出現頻度を取得するステップにおいては、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度を計算するステップにおいては、前記関連元情報および取得された前記各関連先情報のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報のカテゴリの出現頻度の総和、前記関連情報蓄積部に蓄積されたすべての関連先情報のカテゴリの出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関係のカテゴリの出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から出現確率分布を推定し、推定した前記出現確率分布を用いて、前記対象情報のカテゴリの出現確率を前記対象情報の認知度として計算する、付記16から付記23のいずれかに記載の関連情報出力方法。
前記各関連先情報および前記各出現頻度を取得するステップにおいては、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度を計算するステップにおいては、前記関連元情報および取得された前記各関連先情報のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報のカテゴリの出現頻度の総和、前記関連情報蓄積部に蓄積されたすべての関連先情報のカテゴリの出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関係のカテゴリの出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から平均出現頻度を計算し、計算した平均出現頻度と前記対象情報のカテゴリの出現頻度との差を前記対象情報の認知度として計算する、付記16から付記23のいずれかに記載の関連情報出力方法。
前記認知度を計算するステップにおいては、前記対象情報が複数のカテゴリに属している場合には、各前記カテゴリについて計算した前記認知度の平均値、最大値または最小値を前記対象情報の認知度とする、付記24または付記25に記載の関連情報出力方法。
前記各関連先情報および前記各出現頻度を取得するステップにおいては、前記関連元情報を受けて、前記関連元情報に関連する複数の関連先情報と、前記関連元情報および各前記関連先情報の関連内容と、前記関連元情報の出現頻度と、前記各関連先情報の出現頻度と、前記関連元情報および前記各関連先情報の関係の出現頻度と、各前記関連内容の出現頻度とを前記関連情報蓄積部から取得し、
前記認知度を計算するステップにおいては、取得された各前記出現頻度に基づいて、前記関連元情報の認知度、取得された前記各関連先情報の認知度、各前記関係の認知度および取得された前記各関連内容の認知度を計算し、
前記関連先情報を選択するステップにおいては、前記関連元情報の認知度、前記関連先情報の認知度、前記関係の認知度および前記関連内容の認知度のうち、2つ以上の組み合わせに基づいて前記各関連先情報の中から1または複数の前記関連先情報を選択し、選択した前記関連先情報を少なくとも出力する、付記16から付記26のいずれかに記載の関連情報出力方法。
前記関連先情報を選択するステップにおいては、前記各関連先情報の中から、前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度が所定の閾値より高く、かつ前記関連内容の認知度が所定の閾値より低くなる前記関連先情報を選択し、かつ、 前記各関連先情報の中から、前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度が所定の閾値より低く、かつ前記関連内容の認知度が所定の閾値より高くなる前記関連先情報を選択する、付記27に記載の関連情報出力方法。
前記各関連先情報の中から、前記関連元情報の認知度および前記関連先情報の一方の認知度が所定の閾値より高く、他方の認知度が所定の閾値より低く、かつ前記関係の認知度が所定の閾値より低くなる前記関連先情報を選択し、
選択した前記関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度が所定の閾値より高く、かつ前記関係の認知度および前記関連内容の認知度の少なくとも一方が所定の閾値より低くなる前記関連先情報を、選択した前記関連先情報の中で高程度に意外性の高い関連先情報として選択し、かつ、前記関連元情報の認知度および前記関連先情報の一方の認知度が所定の閾値より高く、他方の認知度が所定の閾値より低く、前記関係の認知度が所定の閾値より低く、かつ前記関連内容の認知度が所定の閾値より高くなる前記関連先情報を、選択した前記関連先情報の中で中程度に意外性の高い関連先情報として選択する、付記28に記載の関連情報出力方法。
前記各関連先情報および前記各出現頻度を取得するステップにおいては、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度を計算するステップにおいては、前記認知度を計算すべき前記関連内容に対応する前記関連元情報および前記関連先情報の各々のカテゴリ間の関係について、前記カテゴリ間の関係の出現頻度の総和および前記関連内容の出現頻度に基づいて前記関連内容の出現確率を計算し、計算した前記出現確率を前記関連内容の認知度とする、付記27から付記29のいずれかに記載の関連情報出力方法。
関連元情報に関連する情報を出力するための関連情報出力装置において用いられる関連情報出力プログラムを記録した、コンピュータ読み取り可能な記録媒体であって、
前記関連元情報を受けて、前記関連元情報に関連する複数の関連先情報と、前記関連元情報の出現頻度と、関連する各前記関連先情報の出現頻度と、前記関連元情報および前記各関連先情報の関係の出現頻度とを前記関連情報出力装置の内部または外部に設けられた関連情報蓄積部から取得するステップと、
取得された各前記出現頻度に基づいて、前記関連元情報の認知度、取得された前記各関連先情報の認知度、および各前記関係の認知度を計算するステップと、
前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度のうち、2つ以上の組み合わせに基づいて前記各関連先情報の中から1または複数の前記関連先情報を選択し、選択した前記関連先情報を少なくとも出力するステップとをコンピュータに実行させるための関連情報出力プログラムを記録した、コンピュータ読み取り可能な記録媒体。
前記関連先情報を選択するステップにおいては、前記関連元情報、前記関連先情報および前記関係のうちのいずれか2つを選択し、選択したうちの一方の認知度が所定の閾値より高く、他方の認知度が所定の閾値より低くなる前記関連先情報を前記各関連先情報の中から選択する、付記31に記載のコンピュータ読み取り可能な記録媒体。
前記関連先情報を選択するステップにおいては、前記各関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度の少なくとも一方が所定の閾値より高く、かつ前記関係の認知度が所定の閾値より低くなる前記関連先情報を選択する、付記31または付記32に記載のコンピュータ読み取り可能な記録媒体。
前記関連先情報を選択するステップにおいては、さらに、選択した前記関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度が所定の閾値より高く、かつ前記関係の認知度が所定の閾値より低くなる前記関連先情報を、高程度に意外性の高い関連先情報として選択する、付記32または付記33に記載のコンピュータ読み取り可能な記録媒体。
前記認知度を計算するステップにおいては、前記関連元情報、取得された前記各関連先情報および前記各関係のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報の出現頻度の総和、前記関連情報蓄積部に蓄積されたすべての関連先情報の出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関係の出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から出現確率分布を推定し、推定した前記出現確率分布を用いて、前記対象情報の出現確率を前記対象情報の認知度として計算する、付記31から付記34のいずれかに記載のコンピュータ読み取り可能な記録媒体。
前記認知度を計算するステップにおいては、前記関連元情報、取得された前記各関連先情報および前記各関係のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報の出現頻度の総和、前記関連情報蓄積部に蓄積されたすべての関連先情報の出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関係の出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から平均出現頻度を計算し、計算した前記平均出現頻度と前記対象情報の出現頻度との差を前記対象情報の認知度として計算する、付記31から付記34のいずれかに記載のコンピュータ読み取り可能な記録媒体。
前記各関連先情報および前記各出現頻度を取得するステップにおいては、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度を計算するステップにおいては、前記認知度を計算すべき前記関係である対象関係に対応する前記関連元情報および前記関連先情報の、カテゴリ間の関係の出現頻度の総和から出現確率分布を推定し、推定した前記出現確率分布を用いて、前記カテゴリ間の関係の出現確率を前記対象関係の認知度として計算する、付記31から付記36のいずれかに記載のコンピュータ読み取り可能な記録媒体。
前記認知度を計算するステップにおいては、前記対象関係に対応する前記関連元情報および前記関連先情報の少なくとも一方が複数のカテゴリに属している場合には、各前記カテゴリについて計算した前記認知度の平均値、最大値または最小値を前記対象関係の認知度とする、付記37に記載のコンピュータ読み取り可能な記録媒体。
前記各関連先情報および前記各出現頻度を取得するステップにおいては、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度を計算するステップにおいては、前記関連元情報および取得された前記各関連先情報のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報のカテゴリの出現頻度の総和、前記関連情報蓄積部に蓄積されたすべての関連先情報のカテゴリの出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関係のカテゴリの出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から出現確率分布を推定し、推定した前記出現確率分布を用いて、前記対象情報のカテゴリの出現確率を前記対象情報の認知度として計算する、付記31から付記38のいずれかに記載のコンピュータ読み取り可能な記録媒体。
前記各関連先情報および前記各出現頻度を取得するステップにおいては、さらに、前記
関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情
報蓄積部から取得し、
前記認知度を計算するステップにおいては、前記関連元情報および取得された前記各関連先情報のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報のカテゴリの出現頻度の総和、前記関連情報蓄積部に蓄積されたすべての関連先情報のカテゴリの出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関係のカテゴリの出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から平均出現頻度を計算し、計算した平均出現頻度と前記対象情報のカテゴリの出現頻度との差を前記対象情報の認知度として計算する、
付記31から付記38のいずれかに記載のコンピュータ読み取り可能な記録媒体。
前記認知度を計算するステップにおいては、前記対象情報が複数のカテゴリに属している場合には、各前記カテゴリについて計算した前記認知度の平均値、最大値または最小値を前記対象情報の認知度とする、付記39または付記40に記載のコンピュータ読み取り可能な記録媒体。
前記各関連先情報および前記各出現頻度を取得するステップにおいては、前記関連元情報を受けて、前記関連元情報に関連する複数の関連先情報と、前記関連元情報および各前記関連先情報の関連内容と、前記関連元情報の出現頻度と、前記各関連先情報の出現頻度と、前記関連元情報および前記各関連先情報の関係の出現頻度と、各前記関連内容の出現頻度とを前記関連情報蓄積部から取得し、
前記認知度を計算するステップにおいては、取得された各前記出現頻度に基づいて、前記関連元情報の認知度、取得された前記各関連先情報の認知度、各前記関係の認知度および取得された前記各関連内容の認知度を計算し、
前記関連先情報を選択するステップにおいては、前記関連元情報の認知度、前記関連先情報の認知度、前記関係の認知度および前記関連内容の認知度のうち、2つ以上の組み合わせに基づいて前記各関連先情報の中から1または複数の前記関連先情報を選択し、選択した前記関連先情報を少なくとも出力する、付記31から付記41のいずれかに記載のコンピュータ読み取り可能な記録媒体。
前記関連先情報を選択するステップにおいては、前記各関連先情報の中から、前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度が所定の閾値より高く、かつ前記関連内容の認知度が所定の閾値より低くなる前記関連先情報を選択し、かつ、 前記各関連先情報の中から、前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度が所定の閾値より低く、かつ前記関連内容の認知度が所定の閾値より高くなる前記関連先情報を選択する、付記42に記載のコンピュータ読み取り可能な記録媒体。
前記関連先情報を選択するステップにおいては、さらに、前記各関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度が所定の閾値より高く、かつ前記関係の認知度および前記関連内容の認知度の少なくとも一方が所定の閾値より低くなる前記関連先情報を、選択した前記関連先情報の中で高程度に意外性の高い関連先情報として選択し、
前記各関連先情報の中から、前記関連元情報の認知度および前記関連先情報の一方の認知度が所定の閾値より高く、他方の認知度が所定の閾値より低く、かつ前記関係の認知度が所定の閾値より低くなる前記関連先情報を選択し、
選択した前記関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度が所定の閾値より高く、かつ前記関係の認知度および前記関連内容の認知度の少なくとも一方が所定の閾値より低くなる前記関連先情報を、選択した前記関連先情報の中で高程度に意外性の高い関連先情報として選択し、かつ、前記関連元情報の認知度および前記関連先情報の一方の認知度が所定の閾値より高く、他方の認知度が所定の閾値より低く、前記関係の認知度が所定の閾値より低く、かつ前記関連内容の認知度が所定の閾値より高くなる前記関連先情報を、選択した前記関連先情報の中で中程度に意外性の高い関連先情報として選択する、付記43に記載のコンピュータ読み取り可能な記録媒体。
前記各関連先情報および前記各出現頻度を取得するステップにおいては、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度を計算するステップにおいては、前記認知度を計算すべき前記関連内容に対応する前記関連元情報および前記関連先情報の各々のカテゴリ間の関係について、前記カテゴリ間の関係の出現頻度の総和および前記関連内容の出現頻度に基づいて前記関連内容の出現確率を計算し、計算した前記出現確率を前記関連内容の認知度とする、付記42から付記44のいずれかに記載のコンピュータ読み取り可能な記録媒体。
52 関連情報検索部
53 認知度計算部
54 高関心情報絞込み部
55 出力部
56 関連情報蓄積部
64 高関心情報絞込み部
66 関連情報蓄積部
101 CPU
102 メインメモリ
103 ハードディスク
104 入力インターフェイス
105 表示コントローラ
106 データリーダ/ライタ
107 通信インターフェイス
108 キーボード
109 マウス
110 ディスプレイ
111 記録媒体
121 バス
201 関連情報出力装置
301 入力部
302 位置情報検索部
303 関連情報絞込み部
304 出力部
305 位置情報蓄積部
306 関連情報蓄積部
401 位置情報表示装置
Claims (15)
- 関連元情報に関連する情報を出力するための関連情報出力装置であって、
前記関連元情報を受けて、前記関連元情報に関連する複数の関連先情報と、前記関連元情報の出現頻度と、関連する各前記関連先情報の出現頻度と、前記関連元情報および前記各関連先情報の関係の出現頻度とを前記関連情報出力装置の内部または外部に設けられた関連情報蓄積部から取得するための関連情報検索部と、
取得された各前記出現頻度に基づいて、前記関連元情報の認知度、取得された前記各関連先情報の認知度、および各前記関係の認知度を計算するための認知度計算部と、
前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度のうち、2つ以上の組み合わせに基づいて前記各関連先情報の中から1または複数の前記関連先情報を選択し、選択した前記関連先情報を少なくとも出力するための高関心情報絞込み部とを備える、関連情報出力装置。 - 前記高関心情報絞込み部は、前記関連元情報、前記関連先情報および前記関係のうちのいずれか2つを選択し、選択したうちの一方の認知度が所定の閾値より高く、他方の認知度が所定の閾値より低くなる前記関連先情報を前記各関連先情報の中から選択する、請求の範囲第1項に記載の関連情報出力装置。
- 前記高関心情報絞込み部は、前記各関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度の少なくとも一方が所定の閾値より高く、かつ前記関係の認知度が所定の閾値より低くなる前記関連先情報を選択する、請求の範囲第1項または請求の範囲第2項に記載の関連情報出力装置。
- 前記高関心情報絞込み部は、さらに、選択した前記関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度が所定の閾値より高く、かつ前記関係の認知度が所定の閾値より低くなる前記関連先情報を、高程度に意外性の高い関連先情報として選択する、請求の範囲第2項または請求の範囲第3項に記載の関連情報出力装置。
- 前記認知度計算部は、前記関連元情報、取得された前記各関連先情報および前記各関係のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報の出現頻度の総和、前記関連情報蓄積部に蓄積されたすべての関連先情報の出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関係の出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から出現確率分布を推定し、推定した前記出現確率分布を用いて、前記対象情報の出現確率を前記対象情報の認知度として計算する、請求の範囲第1項から請求の範囲第4項のいずれかに記載の関連情報出力装置。
- 前記認知度計算部は、前記関連元情報、取得された前記各関連先情報および前記各関係のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報の出現頻度の総和、前記関連情報蓄積部に蓄積されたすべての関連先情報の出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関係の出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から平均出現頻度を計算し、計算した前記平均出現頻度と前記対象情報の出現頻度との差を前記対象情報の認知度として計算する、請求の範囲第1項から請求の範囲第4項のいずれかに記載の関連情報出力装置。
- 前記関連情報検索部は、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度計算部は、前記認知度を計算すべき前記関係である対象関係に対応する前記関連元情報および前記関連先情報の、カテゴリ間の関係の出現頻度の総和から出現確率分布を推定し、推定した前記出現確率分布を用いて、前記カテゴリ間の関係の出現確率を前記対象関係の認知度として計算する、請求の範囲第1項から請求の範囲第6項のいずれかに記載の関連情報出力装置。 - 前記関連情報検索部は、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度計算部は、前記関連元情報および取得された前記各関連先情報のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報のカテゴリの出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関連先情報のカテゴリの出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から出現確率分布を推定し、推定した前記出現確率分布を用いて、前記対象情報のカテゴリの出現確率を前記対象情報の認知度として計算する、請求の範囲第1項から請求の範囲第7項のいずれかに記載の関連情報出力装置。 - 前記関連情報検索部は、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度計算部は、前記関連元情報および取得された前記各関連先情報のうちの前記認知度を計算すべき対象情報について、前記関連情報蓄積部に蓄積されたすべての関連元情報のカテゴリの出現頻度の総和、および前記関連情報蓄積部に蓄積されたすべての関連先情報のカテゴリの出現頻度の総和のうち前記対象情報に対応する出現頻度総和を取得し、取得した前記出現頻度総和から平均出現頻度を計算し、計算した平均出現頻度と前記対象情報のカテゴリの出現頻度との差を前記対象情報の認知度として計算する、請求の範囲第1項から請求の範囲第7項のいずれかに記載の関連情報出力装置。 - 前記関連情報検索部は、前記関連元情報を受けて、前記関連元情報に関連する複数の関連先情報と、前記関連元情報および各前記関連先情報の関連内容と、前記関連元情報の出現頻度と、前記各関連先情報の出現頻度と、前記関連元情報および前記各関連先情報の関係の出現頻度と、各前記関連内容の出現頻度とを前記関連情報蓄積部から取得し、
前記認知度計算部は、取得された各前記出現頻度に基づいて、前記関連元情報の認知度、取得された前記各関連先情報の認知度、各前記関係の認知度および取得された前記各関連内容の認知度を計算し、
前記高関心情報絞込み部は、前記関連元情報の認知度、前記関連先情報の認知度、前記関係の認知度および前記関連内容の認知度のうち、2つ以上の組み合わせに基づいて前記各関連先情報の中から1または複数の前記関連先情報を選択し、選択した前記関連先情報を少なくとも出力する、請求の範囲第1項から請求の範囲第9項のいずれかに記載の関連情報出力装置。 - 前記高関心情報絞込み部は、前記各関連先情報の中から、前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度が所定の閾値より高く、かつ前記関連内容の認知度が所定の閾値より低くなる前記関連先情報を選択し、かつ、
前記各関連先情報の中から、前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度が所定の閾値より低く、かつ前記関連内容の認知度が所定の閾値より高くなる前記関連先情報を選択する、請求の範囲第10項に記載の関連情報出力装置。 - 前記高関心情報絞込み部は、さらに、前記各関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度が所定の閾値より高く、かつ前記関係の認知度および前記関連内容の認知度の少なくとも一方が所定の閾値より低くなる前記関連先情報を選択し、かつ、
前記各関連先情報の中から、前記関連元情報の認知度および前記関連先情報の一方の認知度が所定の閾値より高く、他方の認知度が所定の閾値より低く、かつ前記関係の認知度が所定の閾値より低くなる前記関連先情報を選択し、
前記高関心情報絞込み部は、選択した前記関連先情報の中から、前記関連元情報の認知度および前記関連先情報の認知度が所定の閾値より高く、かつ前記関係の認知度および前記関連内容の認知度の少なくとも一方が所定の閾値より低くなる前記関連先情報を、選択した前記関連先情報の中で高程度に意外性の高い関連先情報として選択し、かつ、前記関連元情報の認知度および前記関連先情報の一方の認知度が所定の閾値より高く、他方の認知度が所定の閾値より低く、前記関係の認知度が所定の閾値より低く、かつ前記関連内容の認知度が所定の閾値より高くなる前記関連先情報を、選択した前記関連先情報の中で中程度に意外性の高い関連先情報として選択する、請求の範囲第11項に記載の関連情報出力装置。 - 前記関連情報検索部は、さらに、前記関連元情報のカテゴリ情報および関連する前記各関連先情報のカテゴリ情報を前記関連情報蓄積部から取得し、
前記認知度計算部は、前記認知度を計算すべき前記関連内容に対応する前記関連元情報および前記関連先情報の各々のカテゴリ間の関係について、前記カテゴリ間の関係の出現頻度の総和および前記関連内容の出現頻度に基づいて前記関連内容の出現確率を計算し、計算した前記出現確率を前記関連内容の認知度とする、請求の範囲第10項から請求の範囲第12項のいずれかに記載の関連情報出力装置。 - 関連元情報に関連する情報を出力するための関連情報出力装置における関連情報出力方法であって、
前記関連元情報を受けて、前記関連元情報に関連する複数の関連先情報と、前記関連元情報の出現頻度と、関連する各前記関連先情報の出現頻度と、前記関連元情報および前記各関連先情報の関係の出現頻度とを前記関連情報出力装置の内部または外部に設けられた関連情報蓄積部から取得するステップと、
取得された各前記出現頻度に基づいて、前記関連元情報の認知度、取得された前記各関連先情報の認知度、および各前記関係の認知度を計算するステップと、
前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度のうち、2つ以上の組み合わせに基づいて前記各関連先情報の中から1または複数の前記関連先情報を選択し、選択した前記関連先情報を少なくとも出力するステップとを含む、関連情報出力方法。 - 関連元情報に関連する情報を出力するための関連情報出力装置において用いられる関連情報出力プログラムを記録した、コンピュータ読み取り可能な記録媒体であって、
前記関連元情報を受けて、前記関連元情報に関連する複数の関連先情報と、前記関連元情報の出現頻度と、関連する各前記関連先情報の出現頻度と、前記関連元情報および前記各関連先情報の関係の出現頻度とを前記関連情報出力装置の内部または外部に設けられた関連情報蓄積部から取得するステップと、
取得された各前記出現頻度に基づいて、前記関連元情報の認知度、取得された前記各関連先情報の認知度、および各前記関係の認知度を計算するステップと、
前記関連元情報の認知度、前記関連先情報の認知度および前記関係の認知度のうち、2つ以上の組み合わせに基づいて前記各関連先情報の中から1または複数の前記関連先情報を選択し、選択した前記関連先情報を少なくとも出力するステップとをコンピュータに実行させるための関連情報出力プログラムを記録した、コンピュータ読み取り可能な記録媒体。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/583,805 US9152696B2 (en) | 2010-03-12 | 2010-12-28 | Linkage information output apparatus, linkage information output method and computer-readable recording medium |
JP2012504286A JP5803902B2 (ja) | 2010-03-12 | 2010-12-28 | 関連情報出力装置、関連情報出力方法および関連情報出力プログラム |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2010055699 | 2010-03-12 | ||
JP2010-055699 | 2010-03-12 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2011111284A1 true WO2011111284A1 (ja) | 2011-09-15 |
Family
ID=44563122
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2010/073724 WO2011111284A1 (ja) | 2010-03-12 | 2010-12-28 | 関連情報出力装置、関連情報出力方法およびコンピュータ読み取り可能な記録媒体 |
Country Status (3)
Country | Link |
---|---|
US (1) | US9152696B2 (ja) |
JP (1) | JP5803902B2 (ja) |
WO (1) | WO2011111284A1 (ja) |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008250625A (ja) * | 2007-03-30 | 2008-10-16 | Nomura Research Institute Ltd | 検索システム |
Family Cites Families (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6285999B1 (en) * | 1997-01-10 | 2001-09-04 | The Board Of Trustees Of The Leland Stanford Junior University | Method for node ranking in a linked database |
AU2001261232A1 (en) * | 2000-05-08 | 2001-11-20 | Walker Digital, Llc | Method and system for providing a link in an electronic file being presented to a user |
EP1473639A1 (en) * | 2002-02-04 | 2004-11-03 | Celestar Lexico-Sciences, Inc. | Document knowledge management apparatus and method |
JP3912518B2 (ja) * | 2002-09-10 | 2007-05-09 | ソニー株式会社 | サービス提供システムおよび方法 |
WO2004056047A1 (en) * | 2002-12-13 | 2004-07-01 | Internap Network Services Corporation | Topology aware route control |
USRE44964E1 (en) * | 2003-02-28 | 2014-06-24 | Omnex Systems L.L.C. | Quality information management system |
WO2004080008A1 (ja) * | 2003-03-04 | 2004-09-16 | Fujitsu Limited | 連携情報管理システム |
AU2003209790A1 (en) * | 2003-03-07 | 2004-09-28 | Nokia Corporation | A method and a device for frequency counting |
US20050222989A1 (en) * | 2003-09-30 | 2005-10-06 | Taher Haveliwala | Results based personalization of advertisements in a search engine |
US7617205B2 (en) * | 2005-03-30 | 2009-11-10 | Google Inc. | Estimating confidence for query revision models |
KR100600862B1 (ko) * | 2004-01-30 | 2006-07-14 | 김선권 | 인터넷상의 정보자원에 대한 접근 경로를 체계적으로수집하고 검색하는 방법, 및 이 방법을 실행할 수 있는컴퓨터 프로그램을 수록한 기록매체 |
US20060053135A1 (en) * | 2004-09-03 | 2006-03-09 | Biowisdom Limited | System and method for exploring paths between concepts within multi-relational ontologies |
CN1609859A (zh) * | 2004-11-26 | 2005-04-27 | 孙斌 | 搜索结果聚类的方法 |
US9286387B1 (en) * | 2005-01-14 | 2016-03-15 | Wal-Mart Stores, Inc. | Double iterative flavored rank |
US8176041B1 (en) * | 2005-06-29 | 2012-05-08 | Kosmix Corporation | Delivering search results |
US20070271228A1 (en) * | 2006-05-17 | 2007-11-22 | Laurent Querel | Documentary search procedure in a distributed system |
US20080071797A1 (en) * | 2006-09-15 | 2008-03-20 | Thornton Nathaniel L | System and method to calculate average link growth on search engines for a keyword |
US7996393B1 (en) * | 2006-09-29 | 2011-08-09 | Google Inc. | Keywords associated with document categories |
US20080114753A1 (en) * | 2006-11-15 | 2008-05-15 | Apmath Ltd. | Method and a device for ranking linked documents |
US8429220B2 (en) * | 2007-03-29 | 2013-04-23 | International Business Machines Corporation | Data exchange among data sources |
US8234310B2 (en) * | 2007-12-20 | 2012-07-31 | Pottenger William M | Social networking on a website with topic-based data sharing |
US8306987B2 (en) * | 2008-04-03 | 2012-11-06 | Ofer Ber | System and method for matching search requests and relevant data |
US8010545B2 (en) * | 2008-08-28 | 2011-08-30 | Palo Alto Research Center Incorporated | System and method for providing a topic-directed search |
-
2010
- 2010-12-28 WO PCT/JP2010/073724 patent/WO2011111284A1/ja active Application Filing
- 2010-12-28 US US13/583,805 patent/US9152696B2/en active Active
- 2010-12-28 JP JP2012504286A patent/JP5803902B2/ja active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008250625A (ja) * | 2007-03-30 | 2008-10-16 | Nomura Research Institute Ltd | 検索システム |
Non-Patent Citations (3)
Title |
---|
KAZUSHI NISHIMOTO ET AL.: "A study of a supporting system for generating ideas", ZENKOKU TAIKAI KOEN RONBUNSHU, DAI 46 KAI HEISEI 5 NEN ZENKI (6), INFORMATION PROCESSING SOCIETY OF JAPAN, 1 March 1993 (1993-03-01), pages 283 - 284 * |
KENTARO TORISAWA ET AL.: "Jido Seisei sareta Kensaku Directory 'Torishiki' no Genjo", THE ASSOCIATION FOR NATURAL LANGUAGE PROCESSING DAI 14 KAI NENJI TAIKAI HAPPYO RONBUNSHU, THE ASSOCIATION FOR NATURAL LANGUAGE PROCESSING, 17 March 2008 (2008-03-17), pages 729 - 732 * |
MEGUMI ISHII ET AL.: "Evaluation of the Topic Prompt Extraction Method for a Message Stream", IPSJ SIG NOTES, INFORMATION PROCESSING SOCIETY OF JAPAN, vol. 2005, no. 94, 30 September 2005 (2005-09-30), pages 33 - 40 * |
Also Published As
Publication number | Publication date |
---|---|
US9152696B2 (en) | 2015-10-06 |
JPWO2011111284A1 (ja) | 2013-06-27 |
JP5803902B2 (ja) | 2015-11-04 |
US20130007021A1 (en) | 2013-01-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9418128B2 (en) | Linking documents with entities, actions and applications | |
JP5069285B2 (ja) | ウェブサイトのウェブページのような関連するウェブページの間での有用な情報の伝搬 | |
US9069867B2 (en) | Resource management system, method and program for selecting candidate tag | |
US8244773B2 (en) | Keyword output apparatus and method | |
US20130110839A1 (en) | Constructing an analysis of a document | |
US10152478B2 (en) | Apparatus, system and method for string disambiguation and entity ranking | |
JP2014139807A (ja) | ブラウザ履歴を使用したアクション提案の方法及びシステム | |
Karisani et al. | A query term re-weighting approach using document similarity | |
US8352496B2 (en) | Entity name matching | |
US20110202532A1 (en) | Information sharing system, information sharing method, and information sharing program | |
US8977625B2 (en) | Inference indexing | |
KR20180126577A (ko) | 관련 엔티티 탐색 | |
EP2933734A1 (en) | Method and system for the structural analysis of websites | |
De Nies et al. | Bringing Newsworthiness into the 21st Century. | |
US8799314B2 (en) | System and method for managing information map | |
Kumar | Apache Solr search patterns | |
JP2019109782A (ja) | クエリ生成プログラム、クエリ生成方法およびクエリ生成装置 | |
US10877981B1 (en) | Stopword inclusion for searches | |
US8195458B2 (en) | Open class noun classification | |
JP5803902B2 (ja) | 関連情報出力装置、関連情報出力方法および関連情報出力プログラム | |
JP2011053996A (ja) | Webブラウザ選択方法および装置 | |
KR20220127470A (ko) | 가중치를 이용한 사용자 취향분석 및 공연 리스트업 시스템 | |
KR20100138858A (ko) | 사이트 특성 단어를 이용한 연관 사이트 제공 방법 및 시스템 | |
Trevisiol et al. | Local ranking problem on the browsegraph | |
Tremblay-Beaumont et al. | Jukeblog: A recommender system in the music weblogs |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 10847511 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2012504286 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13583805 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 10847511 Country of ref document: EP Kind code of ref document: A1 |