CN103186619A - Method and device for evaluating search results based on non click operation information - Google Patents

Method and device for evaluating search results based on non click operation information Download PDF

Info

Publication number
CN103186619A
CN103186619A CN2011104574247A CN201110457424A CN103186619A CN 103186619 A CN103186619 A CN 103186619A CN 2011104574247 A CN2011104574247 A CN 2011104574247A CN 201110457424 A CN201110457424 A CN 201110457424A CN 103186619 A CN103186619 A CN 103186619A
Authority
CN
China
Prior art keywords
search results
operation information
user
search
clicking operation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN2011104574247A
Other languages
Chinese (zh)
Other versions
CN103186619B (en
Inventor
曹中宇
王超东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Baidu Netcom Science and Technology Co Ltd
Original Assignee
Beijing Baidu Netcom Science and Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Baidu Netcom Science and Technology Co Ltd filed Critical Beijing Baidu Netcom Science and Technology Co Ltd
Priority to CN201110457424.7A priority Critical patent/CN103186619B/en
Publication of CN103186619A publication Critical patent/CN103186619A/en
Application granted granted Critical
Publication of CN103186619B publication Critical patent/CN103186619B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention aims at providing a method and a device for evaluating search results based on non click operation information. The method comprises the steps as follows: at first, obtaining one or more non click operation information of a user to the search results to be evaluated based on the same inquiry sequence; and then, evaluating the search results according to the non click operation information. Compared with the prior art, the non click operation information of the user to the search results to be evaluated based on the same inquiry sequence is obtained and comprises whether a new inquiry sequence is input, operation of sorting the search results and the like, and the search results are evaluated based on a predetermined evaluation rule, so that search requirements can be included in evaluation standards when the search results are subjected to non click operation by the user, thereby providing a more reliable basis for improving the accuracy of evaluating the search results.

Description

A kind of method and apparatus of the evaluation Search Results based on non-clicking operation information
Technical field
The present invention relates to computer realm, relate in particular to a kind of for the technology of estimating Search Results.
Background technology
In the prior art, evaluation to Search Results mainly depends on the user to the click behavior of Search Results, and do not consider that the user does not need described Search Results is carried out the situation that clicking operation can satisfy its search need in search procedure, for example, the user only can satisfy query demand by the Search Results summary that result of page searching is showed, and need not to click the link of each search result items to enter its corresponding page.Owing to ignore the evaluation of the user not being had clicking operation, thus cause to Search Results estimate not accurate enough.
Therefore, how effectively also to estimate Search Results exactly, become one of present problem demanding prompt solution.
Summary of the invention
The purpose of this invention is to provide a kind of for the method and apparatus of estimating Search Results.
According to an aspect of the present invention, provide a kind of computer implemented for the method for estimating Search Results, this method may further comprise the steps:
A. obtain one or more users to the non-clicking operation information of Search Results based on the same queries sequence to be evaluated;
B. according to described non-clicking operation information, described Search Results is estimated.
According to a further aspect in the invention, also provide a kind of for the equipment of estimating Search Results, this equipment comprises:
The operation information acquisition device is used for obtaining one or more users to the non-clicking operation information of Search Results based on the same queries sequence to be evaluated;
Evaluating apparatus is used for according to described non-clicking operation information, and described Search Results is estimated.
Compared with prior art, the present invention is by obtaining the user to the non-clicking operation information of Search Results based on the same queries sequence to be evaluated, as whether importing new search sequence, search result items being drawn selection operation etc., and based on predetermined evaluation rule this Search Results is estimated, the situation that can satisfy search need when the user is carried out no clicking operation to Search Results comprises to evaluation criterion, thereby provides more reliable foundation for the accuracy of raising Search Results evaluation.
Description of drawings
By reading the detailed description of doing with reference to the following drawings that non-limiting example is done, it is more obvious that other features, objects and advantages of the present invention will become:
Fig. 1 illustrates the equipment synoptic diagram that is used for estimating Search Results according to one aspect of the invention;
Fig. 2 illustrates the method flow diagram that is used for estimating Search Results according to a further aspect of the present invention.
Same or analogous Reference numeral represents same or analogous parts in the accompanying drawing.
Embodiment
Below in conjunction with accompanying drawing the present invention is described in further detail.
Fig. 1 illustrates the equipment synoptic diagram according to the evaluation Search Results of one aspect of the invention.Wherein, valuator device 1 comprises operation information acquisition device 111 and evaluating apparatus 112.
At this, valuator device 1 can be a network equipment or be positioned at wherein functional module, wherein, the network equipment includes but not limited to the cloud that computing machine, network host, single network server, a plurality of webserver collection or a plurality of server constitute, at this, cloud is made of a large amount of computing machines or the webserver based on cloud computing (Cloud Computing), wherein, cloud computing is a kind of of Distributed Calculation, a super virtual machine of being made up of the loosely-coupled computing machine collection of a group; Valuator device 1 also can be subscriber equipment, it includes but not limited to any electronic product that can carry out man-machine interaction by keyboard, telepilot, touch pad or voice-operated device with the user, for example computing machine, smart mobile phone, PDA, game machine or IPTV etc.
The following process of valuator device 1 estimating Search Results with reference to Fig. 1 is described in detail:
Particularly, operation information acquisition device 111 obtains one or more users to the non-clicking operation information of Search Results based on the same queries sequence to be evaluated.
At this, described non-clicking operation information include but not limited to following each:
1) user imports the operation information of new search sequence; For example, after obtaining Search Results, the search input frame of user in result of page searching carries out input operation again, and the new search sequence of this input operation correspondence was with last time search sequence was relevant, for example there is the partial query key word identical in two search sequence, can determines that then the user imports new search sequence under the situation that its search need is not satisfied;
2) user is to the selection operation information of drawing of each search result items of Search Results; Wherein, described stroke of selection operation information includes but not limited to that the user draws the number of times of selection operation to each search result items of Search Results; For example, according to the operation relevant information under the mouse point, as the temporal information under the position coordinates under putting and the point, and the operation relevant information that this mouse discharges when correspondingly discharging mouse, as the position coordinates of release and the temporal information of release, be operating as and once draw selection operation to record these, simultaneously, according to the position coordinates under this point and the position coordinates of release, determine the search result items of this time stroke selection operation correspondence, and according to the temporal information under this point and the temporal information of relieving, determine the duration of this time stroke selection operation;
3) user is to the hover operation information of each search result items of Search Results;
Wherein, described user to the hover operation information of each search result items of described Search Results comprise following at least each:
A) the hover operation number of times of user on each search result items of Search Results; For example, obtain in real time mouse move into Search Results certain search result items region time and shift out this regional time, when the time interval of immigration time and throughput time surpasses the predetermined time interval threshold value, be recorded as hover operation one time, and then the total hover operation number of times of statistics user in this Search Results place result of page searching;
B) duration of the each hover operation of user on each search result items of Search Results; For example, obtain in real time mouse move into Search Results certain search result items region time and shift out this regional time the time interval of record immigration time and throughput time, i.e. duration of hover operation at every turn.
Those skilled in the art will be understood that above-mentioned hover operation information only for giving an example, and other hover operation information existing or that may occur from now on also should be included in the protection domain of the present invention as applicable to the present invention, and are contained in this with way of reference.
Those skilled in the art will be understood that above-mentioned non-clicking operation information only for giving an example, and other non-clicking operation information existing or that may occur from now on also should be included in the protection domain of the present invention as applicable to the present invention, and are contained in this with way of reference.
At this, the described mode of obtaining non-clicking operation information includes but not limited to following arbitrary situation:
1) operation information acquisition device 111 obtains the non-clicking operation information that a user carries out the Search Results based on the same queries sequence to be evaluated by subscriber equipment;
2) operation information acquisition device 111 is answered Event triggered in real time or is received a plurality of users of sending from third party device periodically to the non-clicking operation information of Search Results based on the same queries sequence to be evaluated.
In an example, when valuator device 1 was the network equipment, at first, the user was by the interactive device of subscriber equipment, include but not limited to keyboard, mouse, telepilot, touch pad or hand-written equipment, carry out alternately with browser software or the client software of subscriber equipment.Be example with the mouse, when the user draws selection operation in the result of page searching of the browser software of subscriber equipment, this subscriber equipment obtains the user in real time to stroke selection operation of a certain search result items of Search Results, for example subscriber equipment at first obtains the operation relevant information under the mouse point, as the temporal information under the position coordinates under putting and the point, then when discharging this mouse, obtain the operation relevant information that this mouse discharges, as the position coordinates of release and the temporal information of release, and this time of record is operating as a stroke selection operation, simultaneously, according to the position coordinates under this point and the position coordinates of release, determine the search result items of this time stroke selection operation correspondence, and according to the temporal information under this point and the temporal information of release, determine the duration of this time stroke selection operation; Thereby determine to draw the corresponding non-clicking operation information of selection operation with this user, then this non-clicking operation information is sent to valuator device 1 by the communication mode of arranging; Then, operation information acquisition device 111 receives this non-clicking operation information in real time.
In another example, the application programming interface that operation information acquisition device 111 provides via third party device (API), be subjected to predetermined condition or Event triggered ground or send the request message that receives the non-clicking operation information that a plurality of users carry out the Search Results based on the same queries sequence to be evaluated to this third party device termly, and receive the non-clicking operation information that a plurality of users that this third party device returns in response to this request message carry out the Search Results based on the same queries sequence to be evaluated.In another example, third party device initiatively pushes the non-clicking operation information that a plurality of users carry out the Search Results based on the same queries sequence to be evaluated to valuator device 1, and operation information acquisition device 111 receives the non-clicking operation information that these a plurality of users carry out the Search Results based on the same queries sequence to be evaluated.
Those skilled in the art will be understood that the above-mentioned mode of non-clicking operation information of obtaining is only for giving an example; other existing or modes of obtaining non-clicking operation information that may occur from now on are as applicable to the present invention; also should be included in the protection domain of the present invention, and be contained in this with way of reference.
Then, evaluating apparatus 112 is estimated Search Results according to the non-clicking operation information that operation information acquisition device 111 obtains.
At this, the situation of described Search Results being estimated according to predetermined evaluation rule include but not limited to following at least each:
1) user's new search sequence of input after obtaining Search Results;
For example, if the user is obtaining to re-enter new search sequence behind the Search Results in result of page searching, then evaluating apparatus 112 determines that the opinion rating of this Search Results is low;
2) user selects number of times to surpass predetermined drawing drawing of each search result items of Search Results and selects frequency threshold value;
For example, select frequency threshold value scope maximal value if the user selects number of times to surpass predetermined drawing drawing of each search result items of Search Results, then evaluating apparatus 112 determines that the opinion rating of these Search Results be height;
3) user surpasses the predetermined frequency threshold value of hovering at the number of times that hovers of each search result items of Search Results;
For example, if the user surpasses the predetermined frequency threshold value scope maximal value of hovering at the number of times that hovers of each search result items of Search Results, then evaluating apparatus 112 determines that the opinion rating of this Search Results is high;
4) user surpasses the predetermined duration threshold value of hovering at the duration that hovers of each search result items of Search Results;
For example, if the user surpasses the predetermined duration threshold range maximal value of hovering at the duration that hovers of each search result items of Search Results, then evaluating apparatus 112 determines that the opinion rating of this Search Results is high.
Those skilled in the art will be understood that above-mentioned every predetermined evaluation rule not only can be separately be used for evaluating apparatus and estimates Search Results, and wherein multinomial combination is estimated Search Results for evaluating apparatus.
At this, the evaluation criterion that described Search Results is estimated includes but not limited to following arbitrary form:
A) predetermined opinion rating is high, medium and low as predetermined opinion rating; For example, the user among predetermined a plurality of users 80% or more all imports new search sequence and each search result items is not had operations such as stroke selecting, hover after obtaining based on the Search Results of same queries sequence, and then opinion rating is rudimentary; Again for example, the user among predetermined a plurality of users 80% or more selects number of times to surpass predetermined drawing to drawing of a certain search result items of this Search Results and selects frequency threshold value scope maximal value behind the Search Results that obtains based on the same queries sequence, and then opinion rating is intermediate; Again for example, user among default a plurality of users more than 80% is behind the Search Results that obtains based on the same queries sequence, meet the above-mentioned situation 2 that Search Results is estimated), 3), 4) in more than two or two, and do not import new search sequence, then opinion rating is senior;
B) predetermined evaluation criterion for well/bad, for example, draw operation such as select, hover if the user among a plurality of users more than 80% all imports new search sequence and each search result items do not had after obtaining based on the Search Results of same queries sequence, then evaluation result is bad; Again for example, if the user among a plurality of users more than 80% is behind the Search Results that obtains based on the same queries sequence, meet the above-mentioned situation 2 that Search Results is estimated), 3), 4) in more than two or two, and be the new search sequence of input, then evaluation result is for well.
In an example, non-clicking operation information illustrates among a plurality of users 85% user and re-enter new search sequence in the result of page searching based on the Search Results of same search sequence " sound of chopping wood adventure note ", and operation such as each search result items to this Search Results do not draw choosing, hover, then the opinion rating estimated of 112 pairs of these Search Results of evaluating apparatus is low.
Preferably, also can give weight for above-mentioned predetermined evaluation rule according to predetermined weighting rule, evaluating apparatus 112 be according to this equal weight, and weighting is obtained the corresponding evaluation of estimate of non-clicking operation information with definite opinion rating, thereby realizes Search Results is estimated.
In an example, default weighting rule illustrates:
1) when the situation that Search Results is estimated is: the new search sequence of input after based on the acquisition Search Results of same queries sequence of the user among a plurality of users more than 80%, then the evaluation of estimate of this situation correspondence is-10, and this evaluation of estimate is being used for determining that the weight that the corresponding evaluation of estimate of whole non-clicking operation information has is 0.5
2) when the situation that Search Results is estimated be: a plurality of users select the number of times average to surpass predetermined drawing drawing of each search result items of Search Results and select frequency threshold value scope maximal value 3 times, then the evaluation of estimate of this situation correspondence is 10, and this evaluation of estimate is being used for determining that the weight that the corresponding evaluation of estimate of whole non-clicking operation information has is 0.2
3) a plurality of users surpass the predetermined frequency threshold value scope maximal value 5 times of hovering in the number of times average of hovering of each search result items of Search Results, then the evaluation of estimate of this situation correspondence is 10, and this evaluation of estimate is being used for determining that the weight that the corresponding evaluation of estimate of whole non-clicking operation information has is 0.2
4) a plurality of users surpass the predetermined duration threshold range maximal value 120 seconds of hovering in the duration average of hovering of each search result items of Search Results, then the evaluation of estimate of this situation correspondence is 10, and this evaluation of estimate is being used for determining that the weight that the corresponding evaluation of estimate of whole non-clicking operation information has is 0.1;
Non-clicking operation information illustrates a plurality of users, and in the result of page searching based on the Search Results of same search sequence " Grand National Theatre " each search result items to be drawn selection operation number of times average be 5 times, namely surpass predetermined drawing and select frequency threshold value scope maximal value 3 times, these users number of times average of hovering to each search result items in this result of page searching is 6 times, namely surpass the predetermined frequency threshold value scope maximal value 5 times of hovering, these users in this result of page searching to the duration that hovers of each search result items all above the predetermined duration threshold range maximal value 120 seconds of hovering, and 90% user does not import new search sequence among these users in this result of page searching, then evaluating apparatus 112 evaluation of estimate that corresponding evaluation of estimate weighting is obtained Search Results to these non-clicking operation information is 5 (=10 * 0.2+10 * 0.2+10 * 0.1), greater than predetermined weights scope maximal value 3, determine that opinion rating is senior.
Those skilled in the art will be understood that the above-mentioned mode that Search Results is estimated is only for giving an example; other existing or modes that Search Results is estimated that may occur from now on are as applicable to the present invention; also should be included in the protection domain of the present invention, and be contained in this with way of reference.
Preferably, be to work continuously between operation information acquisition device 111 and the evaluating apparatus 112.Particularly, operation information acquisition device 111 obtains one or more users constantly to the non-clicking operation information of Search Results based on the same queries sequence to be evaluated; Then, evaluating apparatus 112 also constantly according to described non-clicking operation information, is estimated described Search Results.At this, it will be understood by those skilled in the art that " continuing " refer to each device constantly carry out the obtaining of above-mentioned non-clicking operation information, to the evaluation of Search Results, until satisfying predetermined stoppage condition, for example acquisition request device 111 stops obtaining non-clicking operation information in a long time.
Preferably (with reference to Fig. 1), evaluating apparatus 112 is estimated Search Results according to the temporal information of non-clicking operation information and these non-clicking operation information generations.
At this, described non-clicking operation information comprises the temporal information that the user imports the operation information of new search sequence, the selection operation information of drawing to each search result items of Search Results, hover operation information and these inputs, draws choosing, hover operation takes place.
At this, the user can represent it the negating of preceding operation in the back operation, and the evaluation weight of the non-clicking operation that takes place in the back is greater than the evaluation weight in the non-clicking operation of preceding generation thus.
In an example, non-clicking operation information illustrates a certain user and is obtaining not import new search sequence behind the Search Results, simultaneously the search result items A in the Search Results being drawn the time that selection operation and this stroke selection operation finish is 12:01PM, and this user is 12:03PM to the time that the search result items B in these Search Results also draws selection operation and the end of this stroke selection operation; Then evaluating apparatus 112 is according to predetermined weighting rule, give evaluation of estimate 20 to the selection operation of drawing to search result items B that takes place in the back, and weight 0.8, the selection operation of drawing to search result items A is given evaluation of estimate 10, and weight 0.2, then to these non-clicking operation information corresponding weight weighting to obtain weights be 9 (=(20 * 0.8+10 * 0.2)/2), be arranged in predetermined weights scope (6 to 11), determine that opinion rating is middle rank.
Those skilled in the art will be understood that the above-mentioned mode that Search Results is estimated is only for giving an example; other existing or modes that Search Results is estimated that may occur from now on are as applicable to the present invention; also should be included in the protection domain of the present invention, and be contained in this with way of reference.
More preferably (with reference to Fig. 1), evaluating apparatus 112 be according to non-clicking operation information, and in conjunction with the clicking operation information of user to Search Results, this Search Results is estimated.
Particularly, evaluating apparatus 112 can be given weight to non-clicking operation and clicking operation, according to predetermined weighting rule, obtains the evaluation of estimate after the weighting, to determine the opinion rating to Search Results.
In an example, predetermined weighting rule illustrates: the evaluation weight of clicking operation correspondence is 0.6, and the evaluation weight of non-clicking operation is 0.4; The user has clicking operation to each search result items, and the evaluation of estimate of this clicking operation correspondence is 8, simultaneously according to non-clicking operation information determine its corresponding opinion rating for high, its weights value for being 18, then evaluating apparatus 112 calculates and obtains weights 12 (=8 * 0.6+18 * 0.4), 12 greater than predetermined opinion rating threshold range maximal value 11, determines that then opinion rating is senior.
Those skilled in the art will be understood that the above-mentioned mode that Search Results is estimated is only for giving an example; other existing or modes that Search Results is estimated that may occur from now on are as applicable to the present invention; also should be included in the protection domain of the present invention, and be contained in this with way of reference.
At this, one or more users are combined to non-clicking operation and the clicking operation of Search Results based on the same queries sequence to be evaluated, the user all can be comprised to evaluation criterion to satisfy search need and need not click the situation that can satisfy search need by clicking Search Results, further improve the accuracy that Search Results is estimated.
Fig. 2 illustrates the method flow diagram according to the evaluation Search Results of one aspect of the invention.
At this, valuator device 1 can be a network equipment or be positioned at wherein functional module, wherein, the network equipment includes but not limited to the cloud that computing machine, network host, single network server, a plurality of webserver collection or a plurality of server constitute, at this, cloud is made of a large amount of computing machines or the webserver based on cloud computing (Cloud Computing), wherein, cloud computing is a kind of of Distributed Calculation, a super virtual machine of being made up of the loosely-coupled computing machine collection of a group; Valuator device 1 also can be subscriber equipment, it includes but not limited to any electronic product that can carry out man-machine interaction by keyboard, telepilot, touch pad or voice-operated device with the user, for example computing machine, smart mobile phone, PDA, game machine or IPTV etc.
The following process of valuator device 1 estimating Search Results with reference to Fig. 2 is described in detail:
Particularly, in step S1, valuator device 1 obtains one or more users to the non-clicking operation information of Search Results based on the same queries sequence to be evaluated.
At this, described non-clicking operation information include but not limited to following each:
1) user imports the operation information of new search sequence; For example, after obtaining Search Results, the search input frame of user in result of page searching carries out input operation again, and the new search sequence of this input operation correspondence was with last time search sequence was relevant, for example there is the partial query key word identical in two search sequence, can determines that then the user imports new search sequence under the situation that its search need is not satisfied;
2) user is to the selection operation information of drawing of each search result items of Search Results; Wherein, described stroke of selection operation information includes but not limited to that the user draws the number of times of selection operation to each search result items of Search Results; For example, according to the operation relevant information under the mouse point, as the temporal information under the position coordinates under putting and the point, and the operation relevant information that this mouse discharges when correspondingly discharging mouse, as the position coordinates of release and the temporal information of release, be operating as and once draw selection operation to record these, simultaneously, according to the position coordinates under this point and the position coordinates of release, determine the search result items of this time stroke selection operation correspondence, and according to the temporal information under this point and the temporal information of relieving, determine the duration of this time stroke selection operation;
3) user is to the hover operation information of each search result items of Search Results;
Wherein, described user to the hover operation information of each search result items of described Search Results comprise following at least each:
A) the hover operation number of times of user on each search result items of Search Results; For example, obtain in real time mouse move into Search Results certain search result items region time and shift out this regional time, when the time interval of immigration time and throughput time surpasses the predetermined time interval threshold value, be recorded as hover operation one time, and then the total hover operation number of times of statistics user in this Search Results place result of page searching;
B) duration of the each hover operation of user on each search result items of Search Results; For example, obtain in real time mouse move into Search Results certain search result items region time and shift out this regional time the time interval of record immigration time and throughput time, i.e. duration of hover operation at every turn.
Those skilled in the art will be understood that above-mentioned hover operation information only for giving an example, and other hover operation information existing or that may occur from now on also should be included in the protection domain of the present invention as applicable to the present invention, and are contained in this with way of reference.
Those skilled in the art will be understood that above-mentioned non-clicking operation information only for giving an example, and other non-clicking operation information existing or that may occur from now on also should be included in the protection domain of the present invention as applicable to the present invention, and are contained in this with way of reference.
At this, the described mode of obtaining non-clicking operation information includes but not limited to following arbitrary situation:
1) in step S1, valuator device 1 obtains the non-clicking operation information that a user carries out the Search Results based on the same queries sequence to be evaluated by subscriber equipment;
2) in step S1, valuator device 1 is answered Event triggered in real time or is received a plurality of users of sending from third party device periodically to the non-clicking operation information of Search Results based on the same queries sequence to be evaluated.
In an example, when valuator device 1 is the network equipment, at first, the user by with the interactive mode of subscriber equipment, include but not limited to keyboard, mouse, telepilot, touch pad or hand-written equipment, carry out alternately with browser software or the client software of subscriber equipment.Be example with the mouse, when the user draws selection operation in the result of page searching of the browser software of subscriber equipment, this subscriber equipment obtains the user in real time to stroke selection operation of a certain search result items of Search Results, for example subscriber equipment at first obtains the operation relevant information under the mouse point, as the temporal information under the position coordinates under putting and the point, then when discharging this mouse, obtain the operation relevant information that this mouse discharges, as the position coordinates of release and the temporal information of release, and this time of record is operating as a stroke selection operation, simultaneously, according to the position coordinates under this point and the position coordinates of release, determine the search result items of this time stroke selection operation correspondence, and according to the temporal information under this point and the temporal information of release, determine the duration of this time stroke selection operation; Thereby determine to draw the corresponding non-clicking operation information of selection operation with this user, then this non-clicking operation information is sent to valuator device 1 by the communication mode of arranging; Then, in step S1, valuator device 1 receives this non-clicking operation information in real time.
In another example, in step S1, the application programming interface that valuator device 1 provides via third party device (API), be subjected to predetermined condition or Event triggered ground or send the request message that receives the non-clicking operation information that a plurality of users carry out the Search Results based on the same queries sequence to be evaluated to this third party device termly, and receive the non-clicking operation information that a plurality of users that this third party device returns in response to this request message carry out the Search Results based on the same queries sequence to be evaluated.In another example, third party device initiatively pushes the non-clicking operation information that a plurality of users carry out the Search Results based on the same queries sequence to be evaluated to valuator device 1, in step S1, valuator device 1 receives the non-clicking operation information that these a plurality of users carry out the Search Results based on the same queries sequence to be evaluated.
Those skilled in the art will be understood that the above-mentioned mode of non-clicking operation information of obtaining is only for giving an example; other existing or modes of obtaining non-clicking operation information that may occur from now on are as applicable to the present invention; also should be included in the protection domain of the present invention, and be contained in this with way of reference.
Then, in step S2, valuator device 1 is estimated Search Results according to its non-clicking operation information of obtaining in step S1.
At this, the situation of described Search Results being estimated according to predetermined evaluation rule include but not limited to following at least each:
1) user's new search sequence of input after obtaining Search Results;
For example, if the user is obtaining to re-enter new search sequence behind the Search Results in result of page searching, then in step S2, valuator device 1 determines that the opinion rating of this Search Results is low;
2) user selects number of times to surpass predetermined drawing drawing of each search result items of Search Results and selects frequency threshold value;
For example, select frequency threshold value scope maximal value if the user selects number of times to surpass predetermined drawing drawing of each search result items of Search Results, then in step S2, valuator device 1 determines that the opinion rating of this Search Results be height;
3) user surpasses the predetermined frequency threshold value of hovering at the number of times that hovers of each search result items of Search Results;
For example, if the user surpasses the predetermined frequency threshold value scope maximal value of hovering at the number of times that hovers of each search result items of Search Results, then in step S2, valuator device 1 determines that the opinion rating of this Search Results is high;
4) user surpasses the predetermined duration threshold value of hovering at the duration that hovers of each search result items of Search Results;
For example, if the user surpasses the predetermined duration threshold range maximal value of hovering at the duration that hovers of each search result items of Search Results, then in step S2, valuator device 1 determines that the opinion rating of this Search Results is high.
Those skilled in the art will be understood that above-mentioned every predetermined evaluation rule not only can be separately be used for 1 pair of Search Results of valuator device and estimates, and wherein multinomial combination is estimated for 1 pair of Search Results of valuator device.
At this, the evaluation criterion that described Search Results is estimated includes but not limited to following arbitrary form:
A) predetermined opinion rating is high, medium and low as predetermined opinion rating; For example, the user among predetermined a plurality of users 80% or more all imports new search sequence and each search result items is not had operations such as stroke selecting, hover after obtaining based on the Search Results of same queries sequence, and then opinion rating is rudimentary; Again for example, the user among predetermined a plurality of users 80% or more selects number of times to surpass predetermined drawing to drawing of a certain search result items of this Search Results and selects frequency threshold value scope maximal value behind the Search Results that obtains based on the same queries sequence, and then opinion rating is intermediate; Again for example, user among default a plurality of users more than 80% is behind the Search Results that obtains based on the same queries sequence, meet the above-mentioned situation 2 that Search Results is estimated), 3), 4) in more than two or two, and do not import new search sequence, then opinion rating is senior;
B) predetermined evaluation criterion for well/bad, for example, draw operation such as select, hover if the user among a plurality of users more than 80% all imports new search sequence and each search result items do not had after obtaining based on the Search Results of same queries sequence, then evaluation result is bad; Again for example, if the user among a plurality of users more than 80% is behind the Search Results that obtains based on the same queries sequence, meet the above-mentioned situation 2 that Search Results is estimated), 3), 4) in more than two or two, and be the new search sequence of input, then evaluation result is for well.
In an example, non-clicking operation information illustrates among a plurality of users 85% user and re-enter new search sequence in the result of page searching based on the Search Results of same search sequence " sound of chopping wood adventure note ", and operation such as each search result items to this Search Results do not draw choosing, hover, then in step S2, the opinion rating that 1 pair of this Search Results of valuator device is estimated is low.
Preferably, also can give weight for above-mentioned predetermined evaluation rule according to predetermined weighting rule, in step S2, valuator device 1 be according to this equal weight, weighting is obtained the corresponding evaluation of estimate of non-clicking operation information with definite opinion rating, thereby realizes Search Results is estimated.
In an example, default weighting rule illustrates:
1) when the situation that Search Results is estimated is: the new search sequence of input after based on the acquisition Search Results of same queries sequence of the user among a plurality of users more than 80%, then the evaluation of estimate of this situation correspondence is-10, and this evaluation of estimate is being used for determining that the weight that the corresponding evaluation of estimate of whole non-clicking operation information has is 0.5
2) when the situation that Search Results is estimated be: a plurality of users select the number of times average to surpass predetermined drawing drawing of each search result items of Search Results and select frequency threshold value scope maximal value 3 times, then the evaluation of estimate of this situation correspondence is 10, and this evaluation of estimate is being used for determining that the weight that the corresponding evaluation of estimate of whole non-clicking operation information has is 0.2
3) a plurality of users surpass the predetermined frequency threshold value scope maximal value 5 times of hovering in the number of times average of hovering of each search result items of Search Results, then the evaluation of estimate of this situation correspondence is 10, and this evaluation of estimate is being used for determining that the weight that the corresponding evaluation of estimate of whole non-clicking operation information has is 0.2
4) a plurality of users surpass the predetermined duration threshold range maximal value 120 seconds of hovering in the duration average of hovering of each search result items of Search Results, then the evaluation of estimate of this situation correspondence is 10, and this evaluation of estimate is being used for determining that the weight that the corresponding evaluation of estimate of whole non-clicking operation information has is 0.1;
Non-clicking operation information illustrates a plurality of users, and in the result of page searching based on the Search Results of same search sequence " Grand National Theatre " each search result items to be drawn selection operation number of times average be 5 times, namely surpass predetermined drawing and select frequency threshold value scope maximal value 3 times, these users number of times average of hovering to each search result items in this result of page searching is 6 times, namely surpass the predetermined frequency threshold value scope maximal value 5 times of hovering, these users in this result of page searching to the duration that hovers of each search result items all above the predetermined duration threshold range maximal value 120 seconds of hovering, and 90% user does not import new search sequence among these users in this result of page searching, then in step S2, valuator device 1 evaluation of estimate that corresponding evaluation of estimate weighting is obtained Search Results to these non-clicking operation information is 5 (=10 * 0.2+10 * 0.2+10 * 0.1), greater than predetermined weights scope maximal value 3, determine that opinion rating is senior.
Those skilled in the art will be understood that the above-mentioned mode that Search Results is estimated is only for giving an example; other existing or modes that Search Results is estimated that may occur from now on are as applicable to the present invention; also should be included in the protection domain of the present invention, and be contained in this with way of reference.
Preferably, valuator device 1 is to work continuously in step S1 and step S2.Particularly, in step S1, valuator device 1 obtains one or more users constantly to the non-clicking operation information of Search Results based on the same queries sequence to be evaluated; Then, in step S2, valuator device 1 also constantly according to described non-clicking operation information, is estimated described Search Results.At this, it will be understood by those skilled in the art that " continuing " refer to valuator device 1 in each step, constantly carry out the obtaining of above-mentioned non-clicking operation information, to the evaluation of Search Results, until satisfying predetermined stoppage condition, for example valuator device 1 stops obtaining non-clicking operation information in a long time.
Preferably (with reference to Fig. 2), in step S2, valuator device 1 is estimated Search Results according to the temporal information of non-clicking operation information and these non-clicking operation information generations.
At this, described non-clicking operation information comprises the temporal information that the user imports the operation information of new search sequence, the selection operation information of drawing to each search result items of Search Results, hover operation information and these inputs, draws choosing, hover operation takes place.
At this, the user can represent it the negating of preceding operation in the back operation, and the evaluation weight of the non-clicking operation that takes place in the back is greater than the evaluation weight in the non-clicking operation of preceding generation thus.
In an example, non-clicking operation information illustrates a certain user and is obtaining not import new search sequence behind the Search Results, simultaneously the search result items A in the Search Results being drawn the time that selection operation and this stroke selection operation finish is 12:01PM, and this user is 12:03PM to the time that the search result items B in these Search Results also draws selection operation and the end of this stroke selection operation; Then in step S2, valuator device 1 is according to predetermined weighting rule, give evaluation of estimate 20 to the selection operation of drawing to search result items B that takes place in the back, and weight 0.8, the selection operation of drawing to search result items A is given evaluation of estimate 10, and weight 0.2, then to these non-clicking operation information corresponding weight weighting to obtain weights be 9 (=(20 * 0.8+10 * 0.2)/2), be arranged in predetermined weights scope (6 to 11), determine that opinion rating is middle rank.
Those skilled in the art will be understood that the above-mentioned mode that Search Results is estimated is only for giving an example; other existing or modes that Search Results is estimated that may occur from now on are as applicable to the present invention; also should be included in the protection domain of the present invention, and be contained in this with way of reference.
More preferably (with reference to Fig. 2), in step S2, valuator device 1 is according to non-clicking operation information, and in conjunction with the clicking operation information of user to Search Results, this Search Results estimated.
Particularly, in step S2, valuator device 1 can be given weight to non-clicking operation and clicking operation, according to predetermined weighting rule, obtains the evaluation of estimate after the weighting, to determine the opinion rating to Search Results.
In an example, predetermined weighting rule illustrates: the evaluation weight of clicking operation correspondence is 0.6, and the evaluation weight of non-clicking operation is 0.4; The user has clicking operation to each search result items, and the evaluation of estimate of this clicking operation correspondence is 8, simultaneously according to non-clicking operation information determine its corresponding opinion rating for high, its weights value for being 18, then in step S2, valuator device 1 calculates and obtains weights 12 (=8 * 0.6+18 * 0.4), 12 greater than predetermined opinion rating threshold range maximal value 11, determines that then opinion rating is senior.
Those skilled in the art will be understood that the above-mentioned mode that Search Results is estimated is only for giving an example; other existing or modes that Search Results is estimated that may occur from now on are as applicable to the present invention; also should be included in the protection domain of the present invention, and be contained in this with way of reference.
At this, one or more users are combined to non-clicking operation and the clicking operation of Search Results based on the same queries sequence to be evaluated, the user all can be comprised to evaluation criterion to satisfy search need and need not click the situation that can satisfy search need by clicking Search Results, further improve the accuracy that Search Results is estimated.
To those skilled in the art, obviously the invention is not restricted to the details of above-mentioned one exemplary embodiment, and under the situation that does not deviate from spirit of the present invention or essential characteristic, can realize the present invention with other concrete form.Therefore, no matter from which point, all should regard embodiment as exemplary, and be nonrestrictive, scope of the present invention is limited by claims rather than above-mentioned explanation, therefore is intended to be included in the present invention dropping on the implication that is equal to important document of claim and all changes in the scope.Any Reference numeral in the claim should be considered as limit related claim.In addition, obviously other unit or step do not got rid of in " comprising " word, and odd number is not got rid of plural number.A plurality of unit of stating in the device claim or device also can be realized by software or hardware by a unit or device.The first, the second word such as grade is used for representing title, and does not represent any specific order.

Claims (10)

  1. One kind computer implemented for the method for estimating Search Results, this method may further comprise the steps:
    A. obtain one or more users to the non-clicking operation information of Search Results based on the same queries sequence to be evaluated;
    B. according to described non-clicking operation information, described Search Results is estimated.
  2. 2. method according to claim 1, wherein, described non-clicking operation information comprise following at least each:
    -described user imports the operation information of new search sequence;
    -described user is to the selection operation information of drawing of each search result items of described Search Results;
    -described user is to the hover operation information of each search result items of described Search Results.
  3. 3. method according to claim 2, wherein, described user to the hover operation information of each search result items of described Search Results comprise following at least each:
    The hover operation number of times of-user on each search result items of described Search Results;
    The duration of the each hover operation of-user on each search result items of described Search Results.
  4. 4. according to each described method in the claim 1 to 3, wherein, described step b comprises:
    -according to the temporal information of described non-clicking operation information and described non-clicking operation information generation, described Search Results is estimated.
  5. 5. according to each described method in the claim 1 to 4, wherein, described step b comprises:
    -according to described non-clicking operation information, and in conjunction with the clicking operation information of described user to described Search Results, described Search Results is estimated.
  6. 6. equipment of be used for estimating Search Results, this equipment comprises:
    The operation information acquisition device is used for obtaining one or more users to the non-clicking operation information of Search Results based on the same queries sequence to be evaluated;
    Evaluating apparatus is used for according to described non-clicking operation information, and described Search Results is estimated.
  7. 7. equipment according to claim 6, wherein, described non-clicking operation information comprise following at least each:
    -described user imports the operation information of new search sequence;
    -described user is to the selection operation information of drawing of each search result items of described Search Results;
    -described user is to the hover operation information of each search result items of described Search Results.
  8. 8. equipment according to claim 7, wherein, described user to the hover operation information of each search result items of described Search Results comprise following at least each:
    The hover operation number of times of-user on each search result items of described Search Results;
    The duration of the each hover operation of-user on each search result items of described Search Results.
  9. 9. according to each described equipment in the claim 6 to 8, wherein, described evaluating apparatus is used for the temporal information according to described non-clicking operation information and described non-clicking operation information generation, and described Search Results is estimated.
  10. 10. according to each described equipment in the claim 6 to 9, wherein, described evaluating apparatus is used for according to described non-clicking operation information, and in conjunction with the clicking operation information of described user to described Search Results, described Search Results is estimated.
CN201110457424.7A 2011-12-30 2011-12-30 A kind of method and apparatus of the evaluation search result based on non click operation information Active CN103186619B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201110457424.7A CN103186619B (en) 2011-12-30 2011-12-30 A kind of method and apparatus of the evaluation search result based on non click operation information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201110457424.7A CN103186619B (en) 2011-12-30 2011-12-30 A kind of method and apparatus of the evaluation search result based on non click operation information

Publications (2)

Publication Number Publication Date
CN103186619A true CN103186619A (en) 2013-07-03
CN103186619B CN103186619B (en) 2018-08-07

Family

ID=48677788

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201110457424.7A Active CN103186619B (en) 2011-12-30 2011-12-30 A kind of method and apparatus of the evaluation search result based on non click operation information

Country Status (1)

Country Link
CN (1) CN103186619B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105637505A (en) * 2013-05-31 2016-06-01 西洛皮斯公司 Method for searching a database

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6269361B1 (en) * 1999-05-28 2001-07-31 Goto.Com System and method for influencing a position on a search result list generated by a computer network search engine
CN1818908A (en) * 2006-03-16 2006-08-16 董崇军 Feedbakc information use of searcher in search engine
CN101030210A (en) * 2006-10-08 2007-09-05 胡继强 Method for searching sort by user action
CN101454780A (en) * 2006-03-30 2009-06-10 谷歌公司 Method of generating a website profile bases on monitoring user activities
CN102298615A (en) * 2011-07-29 2011-12-28 百度在线网络技术(北京)有限公司 Method for displaying research result realized by computer and equipment

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6269361B1 (en) * 1999-05-28 2001-07-31 Goto.Com System and method for influencing a position on a search result list generated by a computer network search engine
CN1818908A (en) * 2006-03-16 2006-08-16 董崇军 Feedbakc information use of searcher in search engine
CN101454780A (en) * 2006-03-30 2009-06-10 谷歌公司 Method of generating a website profile bases on monitoring user activities
CN101030210A (en) * 2006-10-08 2007-09-05 胡继强 Method for searching sort by user action
CN102298615A (en) * 2011-07-29 2011-12-28 百度在线网络技术(北京)有限公司 Method for displaying research result realized by computer and equipment

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105637505A (en) * 2013-05-31 2016-06-01 西洛皮斯公司 Method for searching a database

Also Published As

Publication number Publication date
CN103186619B (en) 2018-08-07

Similar Documents

Publication Publication Date Title
CN102402619B (en) Search method and device
CN102316130B (en) A kind of behavior based on user judges the method and apparatus of the close and distant degree of itself and good friend
CN102088419B (en) Method and system for searching information of good friends in social network
CN104462293A (en) Search processing method and method and device for generating search result ranking model
CN104077415B (en) Searching method and device
CN104657372A (en) Page operation data processing method and device
CN105302809A (en) Group user level association method and system
CN103838754B (en) Information retrieval device and method
CN103902597A (en) Method and device for determining search relevant categories corresponding to target keywords
CN102497333B (en) In order to export the method for contact object Weather information, client and system
CN106777282B (en) The sort method and device of relevant search
CN109241403A (en) Item recommendation method, device, machinery equipment and computer readable storage medium
CN111310061B (en) Full-link multi-channel attribution method, device, server and storage medium
CN102521257A (en) Method and device for providing corresponding on-line picture according to thumbnail
CN102142035A (en) Method and equipment for sequencing multiple character information records
CN105930527A (en) Searching method and device
CN102222021A (en) Input method panel switching method and device
CN103164425B (en) Method and device for pushing application program in application program platform system
CN102262660A (en) Method and device implemented by computer and used for obtaining search result
CN103186604A (en) Method, device and equipment for determining satisfaction degree of user on search result
US8825698B1 (en) Showing prominent users for information retrieval requests
CN104462347A (en) Keyword classifying method and device
CN202309755U (en) Information service system and information service device
CN103186619A (en) Method and device for evaluating search results based on non click operation information
CN113111216B (en) Advertisement recommendation method, device, equipment and storage medium

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant