CA2575734C - Method and system for identifying keywords for use in placing keyword-targeted advertisements - Google Patents

Method and system for identifying keywords for use in placing keyword-targeted advertisements Download PDF

Info

Publication number
CA2575734C
CA2575734C CA2575734A CA2575734A CA2575734C CA 2575734 C CA2575734 C CA 2575734C CA 2575734 A CA2575734 A CA 2575734A CA 2575734 A CA2575734 A CA 2575734A CA 2575734 C CA2575734 C CA 2575734C
Authority
CA
Canada
Prior art keywords
phrase
documents
words
word
item
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.)
Expired - Fee Related
Application number
CA2575734A
Other languages
French (fr)
Other versions
CA2575734A1 (en
Inventor
Nathaniel B. Scholl
Alexander W. Deneui
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.)
Amazon Technologies Inc
Original Assignee
Amazon Technologies Inc
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 Amazon Technologies Inc filed Critical Amazon Technologies Inc
Publication of CA2575734A1 publication Critical patent/CA2575734A1/en
Application granted granted Critical
Publication of CA2575734C publication Critical patent/CA2575734C/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/31Indexing; Data structures therefor; Storage structures
    • G06F16/313Selection or weighting of terms for indexing
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33Querying
    • G06F16/332Query formulation
    • G06F16/3325Reformulation based on results of preceding query
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241Advertisements
    • G06Q30/0251Targeted advertisements
    • G06Q30/0255Targeted advertisements based on user history
    • G06Q30/0256User search

Abstract

A method and system for identifying search terms for placing advertisements along with search results is provided. The advertisement system selects a description of an item that is to be advertised. The advertisement system then retrieves documents that match the selected description. The advertisement system generates a score for each word of the retrieved documents that indicates relatedness of the word to the item to be advertised. After generating the scores for the words, the advertisement system identifies phrases of the words within the documents that are related to the item. The advertisement system then generates search terms for the item to be advertised from the identified phrases. The advertisement system submits the search terms and an advertisement to a search engines service for placement of a paid-for advertisement for the item.

Description

METHOD AND SYSTEM FOR IDENTIFYING KEYWORDS FOR USE IN PLACING
KEYWORD-TARGETED ADVERTISEMENTS
TECHNICAL FIELD
[0001] The described technology relates generally to terms that are related to an item and specifically to search terms for use in placing advertisements for the item.
BACKGROUND
[0002] Many search engine services, such as Google and Overture, provide for searching for information that is accessible via the Internet. These search engine services allow users to search for web pages and other Internet-accessible resources that may be of interest to users. After a user submits a search request that includes search terms, the search engine service identifies web pages that may be related to those search terms. To quickly identify related web pages, the search engine services may maintain a mapping of keywords to web pages. This mapping may be generated by "crawling" the web (i.e., the World Wide Web) to identify the keywords of each web page. To crawl the web, a search engine service may use a list of root web pages to identify all web pages that are accessible through those root web pages. The keywords of any particular web page can be identified using various well-known information retrieval techniques, such as identifying the words of a headline, the words supplied in the metadata of the web page, the words that are highlighted, and so on. Some search engine services can even search information sources that are not accessible via the Internet. For example, a book publisher may make the content of its books available to a search engine service. The search engine may generate a mapping between the keywords and books. When a search engine service receives a search request that includes one or more search terms, it uses its mapping to identify those information sources (e.g., web pages or books) whose keywords most closely match the search terms. The collection of information sources that most closely matches the search terms is referred to as the "search result." The search engine service then ranks the information sources of the search result based on the closeness of each match, web page popularity (e.g., Google's page ranking), and so on. The search engine service then displays to the user links to those information sources in an order that is based on their rankings.
[0003] Some search engine services do not charge a fee to the providers of web pages for including links to their web pages in search results. Rather, the search engine services obtain revenue by placing advertisements along with search results. These paid-for advertisements are commonly referred to as "sponsored links," "sponsored matches," or "paid-for search results." An advertiser who wants to place an advertisement for an item along with certain search results provides a search engine service with an advertisement and search terms. When a search request is received, the search engine service identifies the advertisements whose search terms most closely match those of the search request. The search engine services can either charge for placement of each advertisement along with search results (i.e., cost per impression) or charge only when a user actually selects a link associated with an advertisement (i.e., cost per click).
[0004] Advertisers would like to maximize the effectiveness of advertising dollars used to pay for advertisements placed along with search results. Those advertisers try to identify search terms for items being advertised that result in the highest benefit (e.g., most profit) to the advertiser. It would be desirable to have techniques that would allow advertisers to maximize the effectiveness of their advertising dollars by identifying search terms that are more targeted to or related to the item being advertised.
BRIEF DESCRIPTION OF THE DRAWINGS
[0005] Figure 1 is a block diagram that illustrates components of the advertisement system in one embodiment.
[0006] Figure 2 is a flow diagram that illustrates the processing of the create advertisement sets component in one embodiment.
[0007] Figure 3 is a flow diagram that illustrates the processing of the score words component in one embodiment.
[0008] Figure 4 is a flow diagram that illustrates the processing of the identify best phrases component in one embodiment.
[0009] Figure 5 is a flow diagram that illustrates the processing of the find phrases component in one embodiment.
[0010] Figure 6 is a flow diagram that illustrates the processing of the score common phrases component in one embodiment.
DETAILED DESCRIPTION
[0011] A method and system for identifying search terms for placing advertisements along with search results is provided. In one embodiment, the advertisement system selects a description of an item that is to be advertised. For example, if the item is a book, then the description may be the title of the book; or if the item is an electronic device, then the description may be a brief summary of the device. The advertisement system then retrieves documents or other information sources that match (e.g., are most closely related to the subject of) the selected description from a corpus of information sources. For example, the advertisement system may submit the selected description to a search engine service with the web pages of the search results being the retrieved documents. The advertisement system then generates a score for each word of the retrieved documents that indicates relatedness of the word to the item to be advertised. In one embodiment, the advertisement system may generate a high score for words that are used much more frequently within the retrieved documents than within the corpus of the information sources. For example, if the item is a Harry Potter book, then words such as "Hogwarts," "Fluffy," "three-headed," "dog," "Hermione," and "Granger"
may have a relatively high score because those words appear more frequently in discussions of Harry Potter than in unrelated discussions. After generating the scores for the words, the advertisement system identifies phrases of the words within the documents that are likely to be related to the item. For example, the advertisement system may identify that the phrases "Fluffy the three-headed dog"
and "Hermione Granger" are likely related to the book. The advertisement system then generates search terms for the item to be advertised from the identified phrases. The advertisement system submits the search terms and an advertisement to a search engines service for placement of a paid-for advertisement for the item.
For example, the advertisement system may place an advertisement for the Harry Potter book with the search term "Hermione Granger." When someone submits a search request of "Hermione Granger" to the search engine service, it will display that advertisement along with the search results. In this way, the advertisement system can identify search terms based on phrases used in information sources that are known to be related to the item that is to be advertised.
[0012] In one embodiment, the advertisement system identifies phrases that are likely to be related to the item to be advertised. Because the number of phrases within a document is 0(n2) when n is the number of words within a document and the number of possible phrases in a corpus of documents is 1c1 where k is the number of distinct words and us the length of the phrase, it would be computationally very expensive to evaluate and track every possible phrase. To reduce the number of phrases that are evaluated, the advertisement system evaluates groups of words with high scores that are in close proximity to each other. The advertisement system initially generates a score for the words within documents that are related to the item.
The score indicates the likelihood that the word is related to the item. The advertisement system may then identify highly related words and related words.
A
highly related word has a very high score such as a score in the top 10%, and a related word has a high score such as a score in the top 25%. The advertisement system searches the documents for the highly related words. Each highly related word within a document is considered the "anchor word" of a phrase. The advertisement system tries to extend the phrase to include nearby related words. In one embodiment, the advertisement system may extend the phrase by any contiguous related words that follow the anchor word. For example, if "Hermione" is a highly related word and "Granger" is a related word, the phrase "Hermione Granger" would be identified as a phrase when "Hermione" is followed by "Granger"
in a document. Alternatively, the advertisement system may extend the phrase to also include words before the anchor word. For example, if "Granger" is a highly related word and "Hermione" is only a related word, then the phrase "Hermione Granger" would still be identified. The advertisement system may calculate a phrase score and continue extending a phrase so long as the score of the extended phrase increases regardless of whether all the words of the phrase are related words.
One skilled in the art will appreciate that the technique for identifying such phrases may be used in contexts other than generating search terms for advertisements. For example, a search engine service may use the phrases identified in the search results as search requests for locating additional related information sources to present to a user. Alternatively, the advertisement system could identify more phrases from the additional related information sources. More generally, given a corpus of information sources, the technique for identifying phrases can be used to identify topics of the information sources. For example, if the information sources are chat discussions, then the identified phrases may represent the most popular topics of the chat discussions.
[0013] Figure 1 is a block diagram that illustrates components of the advertisement system in one embodiment. The advertisement system 110 is connected to search engine service computer systems 101 and web server computer systems 102 via a communications link 103. The advertisement system submits a description of an item to a search engine service computer system and receives links to matching web pages that are provided by the web server computer systems. The advertisement system then retrieves the matching web pages from the web server computer systems. The advertisement system identifies phrases from those matching web pages and derives search terms from the identified phrases. The advertisement system then submits to the search engine services the search terms along with an advertisement for the item. The search engine services display the advertisement along with search results for a search query that matches the search terms.
[0014] The advertisement system includes a create advertisement sets component 111, a score words component 112, an identify best phrases component 113, a find phrases component 114, a score common phrases component 115, an item data store 116, a search results store 117, and a score store 118. The item data store contains an identifier (e.g., SKU) of each item to be advertised along with a description of the item. For example, the item data store may be an electronic catalog of books that are to be advertised. Each, catalog entry may include an item identifier, a title, an author name, a summary, and so on. The search results store contains the matching web pages for the item for which search terms are being identified. The score store contains the score for the words and phrases of the search results store. The create advertisement sets component is provided with an item identifier and identifies search terms (e.g., keywords) to be used when advertising that item. The create advertisement sets component requests a search engine service to provide search results, retrieves the web pages of those search results, invokes the score words component and the identify best phrases component, and then generates the advertisement sets. The score words component generates a score for each word of the search results that indicates a likelihood that the word is related to the item. The identify best phrases component invokes the find phrases component and the score common phrases component to identify phrases that are likely to be related to the item.
[0015] The advertisement system may be implemented on computer systems and servers that include a central processing unit, a memory, input devices (e.g., keyboard and pointing devices), output devices (e.g., display devices), and storage devices (e.g., disk drives). The memory and storage devices are computer-readable media that may contain instructions that implement the advertisement system.
In addition, the data structures and message structures may be stored or transmitted via a data transmission medium, such as a signal on a communications link.
Various communications links may be used, such as the Internet, a local area network, a wide area network, or a point-to-point dial-up connection.
[0016] Figure 2 is a flow diagram that illustrates the processing of the create advertisement sets component in one embodiment. The component is passed an identifier of an item and returns the advertisement sets with search terms derived from phrases that are likely to be related to the item. In block 201, the item retrieves a description of the item. For example, the description may be the title of the book or the item name combined with the manufacturer name (e.g., "Sony DVD player").
In block 202, the component requests a search engine service to perform a search using the retrieved description as the search request. The component receives the search results. If the search results are links, such as URLs to web pages, then the component retrieves the linked web pages and stores them in the search results store. The component may store and use only the best matching web pages (e.g., the top 15) of the search results. In block 203, the component invokes the score words component to generate a score for each word in the search results. The invoked component stores the scores in the score store. In block 204, the component invokes the identify best phrases component to identify the phrases that are most highly related to the item. The invoked component stores the phrase scores in the score store. In block 205, the component generates advertisement sets for the item using the best phrases. The component then completes. These advertisment sets may then be submitted to one or more search engine services.
[0017] Figure 3 is a flow diagram that illustrates the processing of the score words component in one embodiment. The score words component generates a score for each word stored in the web pages of the search results store. The component stores the scores in the score store. In blocks 301-308, the component loops selecting each word in the search results and calculating its score. In block 301, the component selects the next word in the search results. In decision block 302, if all the words in the search results have already been selected, then the component returns, else the component continues at block 303. One skilled in the art will appreciate that the component may skip noise words (e.g., "of," "a,"
"the," and so on). In block 303, the component calculates the average frequency of the selected word within the documents (e.g., web pages) of the search results.
The "frequency" of a word is the number of occurrences of that word within the document divided by the total number of occurrences of words within that document. For example, if a word occurs 10 times within a document that contains 200 words, then its frequency is .05 (i.e., 10/200), which means that it accounts for 5% of the words in the document. The "average frequency" of a word within the search results is the average of the frequencies of that word for each document. For example, if the frequencies for a word are .05, .04, .02, and .01 in a search result that has four documents, then the average frequency for that word is .03 (e.g., (.05+.04+.02+.01)/4). The average frequency is represented by the following equation:
7 _ _______________________________________________________________ (1) where 7 is the average frequency of a word, fi is the frequency of the word in document i, and n is the number of documents. In block 304, the component retrieves the "normal frequency" for the word. The normal frequency represents the average frequency of the word in a very large corpus of documents, such as all web pages. In block 305, the component calculates a "frequency score" for the selected word. If the average frequency of the selected word is much higher than the normal frequency of the selected word, then the word may be highly related to the item. The frequency score provides a scoring of the average frequency relative to the normal frequency. The frequency score may be represented by the following equation:
= +
atan (i ¨ I
10 * f) (2) Si .5 where Sf is the frequency score for the word, f is the normal frequency of the word, and atan is the arc tangent function. One skilled in the art will appreciate that this equation is just one of many equations that can be used to generate the frequency score. The particular equation used can be selected based on the weight to be given to the difference between the average and normal frequencies of a word.
In block 306, the component calculates the number of documents of the search results that contain the selected word. In block 307, the component calculates a "contain score" that indicates the fraction of the documents of the search results that contain the selected word. The contain score may be represented by the following equation:
n' (3) S = ¨
C n where Sc is the contain score and n' is the number of documents of the search results that contain the selected word. In block 308, the component calculates the score for the selected word. In one embodiment, the word score is a linear combination of the frequency score and the contain score. The weight of the frequency score and the contain score can be set to reflect whether the frequency score or the contain score is considered to be a more accurate representation of the likelihood that the word is related to the item. The word score may be represented by the following equation:
S=a*Sf+ (1 ¨ a) * S c (4) where S is the word score and a varies from zero to one and represents the weight given to the frequency score. The component then loops to block 301 to select the next word in the search results.
[0018]
Figure 4 is a flow diagram that illustrates the processing of the identify best phrases component in one embodiment. In block 401, the component selects the highly related words of the search results. The highly related words may be those words whose score is in the top 15%. The highly related words are used as the anchor words for the phrases. In block 402, the component selects the related words of the search results. The related words may be those words whose score is in the top 40%. The related words include the highly related words. The phrase may be extended to include related words that are near the anchor word. One skilled in the art will appreciate that various criteria can be used to select the highly related words and the related words. For example, the highly related words might be the 10 words with the top scores, and the related words might be the 50 words with the top scores. In addition, the highly related words and the related words could be the same set of words (e.g., the 20 words with the top scores). In blocks 403-405, the component loops selecting documents in the search results and finding phrases within those documents. In block 403, the component selects the next document in the search results. In decision block 404, if all the documents in the search results have already been selected, then the component continues at block 406, else the component continues at block 405. In block 405, the component invokes the find phrases component to find the phrases within the selected document. The component then loops to block 403 to select the next document. In block 406, after the phrases have been found in all the documents, the component selects common phrases, that is, phrases that occur frequently within the documents. For example, a common phrase may be one that occurs more than five times within the documents or that occurs in a certain percentage of the documents. In block 407, the component invokes the score common phrases component to generate a phrase score for each common phrase. The component then returns. The advertisement system derives the search terms from the common phrases.
[0019] Figure 5 is a flow diagram that illustrates the processing of the find phrases component in one embodiment. This component is passed a document and identifies the phrases within the document. In blocks 501-509, the component loops identifying phrases within the documents that have highly related words as anchor words. In block 501, the component selects the next highly related word within the document. In decision block 502, if all the highly related words of the document have already been selected, then the component completes, else the component continues at block 503. In block 503, the component initializes the phrase with the selected highly related word as the anchor word. In blocks 504-509, the component loops extending the phrase to include related words that are nearby. In block 504, the component selects the next word within the document. In decision block 505, if the selected word is a related word, then the component continues at block 506, else the component terminates the extending of the phrase and loops to block 501 to identify the next phrase within the document. In decision block 506, if the selected word is similar to a word already in the phrase, then the component terminates the extending of the phrase and loops to block 501 to identify the next phrase, else the component continues at block 507. In decision block 507, if the selected word will improve the phrase score, then the component continues at block 509, else the component continues at block 508. In decision block 508, if the selected word and the next word after the selected word would improve the phrase score, then the component continues at block 509, else the component terminates the extending of the phrase and loops to block 501 to identify the next phrase. In block 509, the component adds the selected word to the phrase and loops to block 504 to select the next word for extending the phrase.
[0020] Figure 6 is a flow diagram that illustrates the processing of the score common phrases component in one embodiment. The component calculates a phrase score for the common phrases. Alternatively, the phrase scores may be calculated as each common phrase is identified. In block 601, the component selects the next common phrase. In decision block 602, if all the common phrases have already been selected, then the component returns, else the component continues at block 603. In block 603, the component initializes the phrase score for the selected common phrase. In blocks 604-607, the component loops factoring in the word scores of the words of the common phrase into the phrase score. In block 604, the component selects the next word of the selected common phrase. In decision block 605, if all the words of the selected common phrase have already been selected, then the component continues at block 607, else the component continues at block 606. In block 606, the component adds the word score of the selected word to the phrase score and then loops to block 604 to select the next word of the selected common phrase. One skilled in the art will appreciate that many different techniques may be used for calculating a phrase score. For example, double the word score of highly related words may be added to the phrase score to =
emphasise the importance of highly related words, a nonlinear combination of word scores may be used, and so on. In block 607, the component multiplies the phrase score by the number of occurrences of the selected common phrase within the search results and the component then loops to block 601 to select the next common phrase.
[0021]
One skilled in the art will appreciate that although specific embodiments of the advertisement system have been described herein for purposes of illustration, various modifications may be made. The term "item" includes any product, service, or concept that can be advertised. For example, a political party can place advertisements relating to a particular candidate or cause. In addition, an advertisement set may not have a link associated with it. An advertiser may want to simply display the information of an advertisement to users who submit requests using a certain search term. For example, a candidate may want an advertisement displayed when a user submits a search request with the name of their opponent as a search term. One skilled in the art will appreciate that various equations and techniques for calculating scores can be used. Also, if the search results contain documents that are duplicates (or very similar), the advertising system may disregard the duplicate documents. The advertisement system may maintain a list of words that should not be added to phrases, such as a word that is very common on all web pages (e.g., "next page" or "privacy policy").
Accordingly, the invention is not limited except by the appended claims.

Claims (102)

I/We claim:
1. A computer-implemented method for identifying phrases related to an item from documents related to the item, the method comprising:
by a computer, generating a score for each of a plurality of words of the documents, the score indicating relatedness of a corresponding word to the item, wherein the plurality of words includes a first subset of words and a second subset of words, the first subset of words having top scores among a first top percentage of the plurality of words, the second subset of words having top scores among a second top percentage of the plurality of words, the second top percentage being larger than the first top percentage;
by the computer, selecting at least one word from the first subset of words;
by the computer, locating each of the at least one word within the documents as an anchor word of a phrase; and by the computer, extending the phrase by one or more words proximate to the phrase if the one or more words are in the second subset of words.
2. The method of claim 1, further comprising:
selecting a description of the item; and selecting as related to the item documents that match the selected description of the item.
3. The method of claim 2 wherein the selection of documents related to the item includes submitting the selected description of the item to a search engine service and selecting documents based on search results provided by the search engine service.
4. The method of claim 2 wherein the selecting of the description includes retrieving the description from an item catalog.
5. The method of claim 4 wherein the description is a title of the item that is stored in the item catalog.
6. The method of claim 1 wherein generating the score for a first word of the documents comprises:
calculating the average frequency of the first word by applying the formula:
where .function. is the average frequency of the first word, .function. i is the frequency of the first word in document i, and .eta. is the number of documents, and retrieving the normal frequency .function. of the first word, wherein the normal frequency represents the average frequency of the first word in a very large corpus of documents.
7. The method of claim 6, further comprising calculating a frequency score of the first word by applying the formula:
where S .function. is the frequency score of the first word.
8. The method of claim 7, further comprising calculating a contain score of the first word by applying the formula:
where S c is the contain score and .eta.' is the number of documents that contain the first word.
9. The method of claim 8 wherein generating the score of the first word comprises applying the formula:
S = .alpha. * S .function. + (1¨ .alpha.) * S c where S is the score of the first word and .alpha. is a constant between zero and one representing the weight given to the frequency score.
10. The method of claim 1, further comprising generating scores for phrases wherein a phrase is extended when extending would result in a score indicating that the extended phrase is more related than the unextended phrase to the item.
11. The method of claim 1 wherein a phrase is only extended by words with scores indicating a relatedness to the item.
12. The method of claim 1 wherein relatedness of a phrase to the item is determined to be high based on a number of occurrences of the phrase within the documents.
13. The method of claim 1 wherein a word that is similar to another word in a phrase is not added to the phrase.
14. The method of claim 1 wherein a phrase is ended when a word that is similar to a word already in the phrase is encountered.
15. The method of claim 1 wherein noise words are ignored.
16. The method of claim 1 wherein words that generally score highly in a general corpus of documents are ignored.
17. The method of claim 1 wherein documents that are similar to other retrieved documents are ignored.
18. The method of claim 1, further comprising placing an advertisement for the item with at least one search term that is the same as an extended phrase.
19. The method of claim 1, further comprising displaying an advertising message for the item to a user who has submitted a query containing a phrase among the extended phrases.
20. A computer-readable medium containing computer-executable instructions thereon that, when executed by a computer, cause the computer to identify phrases related to an item from information sources related to the item, by a method comprising:
generating a score for words of the information sources, a generated score indicating relatedness of the word to the item;
locating within the information sources words with top scores as an anchor word of a phrase; and extending each phrase starting with the anchor word of the phrase by words proximate to the phrase based on relatedness of the extended phrase to the item.
21. The computer-readable medium of claim 20, the method further comprising:
selecting a description of the item; and selecting as related to the item documents that match the selected description of the item.
22. The computer-readable medium of claim 21 wherein the selection of documents related to the item includes submitting the selected description of the item to a search engine service and selecting documents based on search results provided by the search engine service.
23. The computer-readable medium of claim 21 wherein the selecting of the description includes retrieving the description from an item catalog.
24. The computer-readable medium of claim 23 wherein the description is a title of the item that is stored in the item catalog.
25. The computer-readable medium of claim 20, the method further comprising generating scores for phrases wherein a phrase is extended when extending would result in a score indicating that the extended phrase is more related, than the unextended phrase, to the item.
26. The computer-readable medium of claim 20 wherein a phrase is only extended by words with scores indicating a relatedness to the item.
27. The computer-readable medium of claim 20 wherein relatedness of a phrase to the item is determined to be high based on a number of occurrences of the phrase within the information sources.
28. The computer-readable medium of claim 20 wherein a word that is similar to another word in a phrase is not added to the phrase.
29. The computer-readable medium of claim 20 wherein a phrase is ended when a word that is similar to a word already in the phrase is encountered.
30. The computer-readable medium of claim 20 wherein words that generally score highly in a general corpus of information sources are ignored.
31. The computer-readable medium of claim 20 wherein documents that are similar to other retrieved documents are ignored,
32. The computer-readable medium of claim 20, the method further comprising placing an advertisement for the item with at least one search term that is the same as an extended phrase.
33. The computer-readable medium of claim 20, the method further comprising displaying an advertising message for the item to a user who has submitted a query containing a phrase among the extended phrases.
34. A system for identifying phrases related to an item from information sources related to the item, the system comprising a computer, the computer comprising:

a scoring subsystem that generates a score for words of the information sources, each generated score indicating relatedness of the word to the item;
a location subsystem that locates within the information sources words with top scores each as an anchor word of a phrase; and a phrase extension subsystem that extends each phrase starting with its anchor word by words proximate to the phrase based on relatedness of the extended phrase to the item.
35. An apparatus for identifying phrases related to an item from documents related to the item, the apparatus comprising:
a computer-readable medium containing computer-executable instructions thereon; and a central processing unit configured to execute the computer-executable instructions;
wherein the computer-executable instructions are configured to cause the central processing unit to execute the method according to any one of claims 1 to 19.
36. An apparatus for scoring phrases related to an item from documents related to the item, the apparatus comprising:
a data store;
a computer-readable medium containing computer-executable instructions thereon; and a central processing unit configured to execute the computer-executable instructions;
wherein the computer-executable instructions are configured to cause the central processing unit to:
generate a word score for words of the documents, the word score indicating relatedness of the word to the item;
save the word scores to the data store;
locate each selected word within the document as an anchor word of a phrase;

extend each phrase by words proximate to the phrase based on relatedness of the extended phrase to the item;
generate a phrase score for each phrase; and save the phrase scores to the data store.
37. A computer-implemented method for identifying phrases related to an item, the method comprising:
by a computer, submitting a description of the item to a search engine service, by the computer, retrieving search result documents from the search engine service, by the computer, generating a score for words of the documents, the score indicating relatedness of the word to the item;
by the computer, selecting words with top scores;
by the computer, locating each selected word within the documents as an anchor word of a phrase; and by the computer, extending each phrase by words proximate to the phrase based on relatedness of the extended phrase to the item.
38. A computer-implemented method for searching a corpus of documents for relevant documents, the method comprising:
by a computer, processing a plurality of documents retrieved from a search engine service in response to a search request, to identify one or more phrases relevant to the search results, the processing comprising:
scoring words in the plurality of documents based on one or more of:
average frequencies of the words in the documents;
average frequencies of the words in the corpus of documents; and a fraction of the plurality of documents that contain the words;

based on the scoring, identifying the one or more relevant phrases from the plurality of documents wherein the identified phrases incorporate words in a top percentile of the scores;
applying the one or more relevant phrases as search requests to the search engine service for locating additional relevant information to present to a user.
39. The method of claim 38 wherein scoring the words comprises:
calculating the average frequency of one of the words by applying the formula:
where ~ is the average frequency of the one of the words, fi is the frequency of the one of the words in document i, and n is the number of documents in the plurality of documents, and retrieving the normal frequency ~ of the one of the words, wherein the normal frequency represents the average frequency of the one of the words in the corpus of documents.
40. The method of claim 39, further comprising calculating a frequency score of the one of the words by applying the formula:
where Sf is the frequency score of the one of the words.
41. The method of claim 40, further comprising calculating a contain score of the one of the words by applying the formula:
where Sc is the contain score and n' is the number of documents in the plurality of documents that contain the one of the words.
42. The method of claim 41 wherein generating the score of the one of the words comprises applying the formula:
S = .alpha. * S .function. + (1 ¨ .alpha.) * Sc where S is the score of the one of the words and a is a constant between zero and one representing the weight given to the frequency score.
43. The method of any one of claims 38 to 42 wherein, identifying the one or more relevant phrases comprises selecting one of the words in the top percentile of scores as an anchor word of one of the one or more phrases and by the computer, extending the one of the one or more phrases by one or more additional words proximate to the one of the one or more phrases if the one or more additional words have scores among a second top percentile of the plurality of words.
44. The method of claim 43, further comprising generating scores for the one or more phrases wherein a first phrase is extended when extending would result in a score indicating that the extended first phrase is more related than the unextended first phrase to a subject of the search request.
45. The method of claim 38 comprising determining a number of occurrences of the one or more phrases within the documents.
46. The method of claim 43 wherein a first word that is similar to another word in a specific one of the one or more phrases is not added to the specific phrase.
47. The method of claim 43 wherein a specific phrase is ended when a second word that is similar to another word already in the phrase is encountered.
48. The method of any one of claims 38 to 47 wherein noise words are ignored.
49. The method of any one of claims 38 to 48 wherein words that generally score highly in the corpus of documents are ignored.
50. The method of any one of claims 38 to 49 wherein documents that are similar to other retrieved documents are ignored.
51. A computer-readable medium containing computer-executable instructions thereon that, when executed by a computer, cause the computer to perform a method according to any one of claims 38 to 50.
52. A computer-implemented method for identifying phrases related to an item from documents related to the item, the method comprising:
by a computer, generating a score for each of a plurality of words of the documents, the score indicating relatedness of a corresponding word to the item, wherein generating the score is based on one or more of:
average frequencies of the words in the documents;
average frequencies of the words in a corpus of documents;
and a fraction of the documents that contain the words;
based on the scoring, identifying one or more phrases relevant to the item from the documents wherein the identified phrases incorporate words in a top percentile of the scores.
53. The method of claim 52, further comprising:
selecting a description of the item; and selecting, as related to the item, documents from the corpus of documents that match the selected description of the item.
54. The method of claim 53 wherein the selection of documents related to the item includes submitting the selected description of the item to a search engine service and selecting documents based on search results provided by the search engine service.
55. The method of claim 53 wherein the selecting of the description includes retrieving the description from an item catalog.
56. The method of claim 55 wherein the description is a title of the item that is stored in the item catalog.
57. The method of any one of claims 52 to 56 wherein generating the score for a first word of the documents comprises:
calculating the average frequency of the first word by applying the formula:
where ~ is the average frequency of the first word, .function.i is the frequency of the first word in document i, and n is the number of the documents, and retrieving the normal frequency ~ of the first word, wherein the normal frequency represents the average frequency of the first word in the corpus of documents.
58. The method of claim 57, further comprising calculating a frequency score of the first word by applying the formula:
where Sf is the frequency score of the first word.
59. The method of claim 58, further comprising calculating a contain score of the first word by applying the formula:
where Sc is the contain score and n' is the number of documents that contain the first word.
60. The method of claim 59 wherein generating the score of the first word comprises applying the formula:
S = .alpha. * S .function. + (1 ¨ .alpha.) * Sc where S is the score of the first word and a is a constant between zero and one representing the weight given to the frequency score.
61. The method of claim 52, further comprising generating scores for phrases wherein a phrase is extended when extending would result in a score indicating that the extended phrase is more related than the unextended phrase to the item.
62. The method of claim 61 wherein a phrase is only extended by words with scores indicating a relatedness to the item.
63. The method of claim 61 wherein relatedness of a phrase to the item is determined to be high based on a number of occurrences of the phrase within the documents.
64. The method of any one of claims 61 to 63 wherein a word that is similar to another word in a phrase is not added to the phrase.
65. The method of any one of claims 61 to 64 wherein a phrase is ended when a word that is similar to a word already in the phrase is encountered.
66. The method of any one of claims 52 to 65 wherein noise words are ignored.
67. The method of any one of claims 52 to 66 wherein words that generally score highly in a general corpus of documents are ignored.
68. The method of any one of claims 52 to 67 wherein documents that are similar to other ones of the documents are ignored.
69. The method of any one of claims 52 to 68, further comprising placing an advertisement for the item with at least one search term that is the same as an extended phrase.
70. The method of any one of claims 52 to 69, further comprising automatically displaying an advertising message for the item to a user who has submitted a query containing a phrase among the extended phrases.
71. A system for searching a corpus of documents for relevant documents, the system comprising a computer, the computer comprising:
a processor for processing a plurality of documents retrieved from a search engine service in response to a search request, the processor implementing:
a scoring generator for scoring words in the documents based on one or more of:
average frequencies of the words in the documents;
average frequencies of the words in the corpus of documents; and a fraction of the documents that contain the words;
a phrase identifier for identifying one or more phrases from the documents relevant to the search request based on the scoring, wherein the identified phrases incorporate words in a top percentile of the scores; and a search expander for applying the one or more phrases as search requests to the search engine service for locating additional relevant information to present to a user.
72. The system of claim 71 wherein the scoring generator is configured to score the words by:
calculating the average frequency of one of the words by applying the formula:
where ~ is the average frequency of the one of the words, .function.i is the frequency of the one of the words in document i, and n is the number of documents in the plurality of documents, and retrieving the normal frequency ~ of the one of the words, wherein the normal frequency represents the average frequency of the one of the words in the corpus of documents.
73. The system of claim 72, wherein the scoring generator is configured to score the words by calculating a frequency score of the one of the words by applying the formula:
where S .function. is the frequency score of the one of the words.
74. The system of claim 73, wherein the scoring generator is configured to score the words by calculating a contain score of the one of the words by applying the formula:
where Sc is the contain score and n' is the number of documents in the plurality of documents that contain the one of the words.
75. The system of claim 74 wherein the scoring generator is configured to score the words by applying the formula:
S = a * S .function. + (1¨ .alpha.)* Sc where S is the score of the one of the words and a is a constant between zero and one representing the weight given to the frequency score.
76. The system of any one of claims 71 to 75 wherein the processor is configured to select one of the words in the top percentile of scores as an anchor word of one of the one or more phrases and to extend the one of the one or more phrases by one or more additional words proximate to the one of the one or more phrases if the one or more additional words have scores among a second top percentile of the plurality of words.
77. The system of claim 76 wherein the phrase identifier is configured for extending a first phrase when extending would result in a score indicating that the extended first phrase is more related than the unextended first phrase to a subject of the search request.
78. The system of claim 71 wherein the computer determines a number of occurrences of the one or more phrases within the documents.
79. The system of claim 76 wherein the phrase identifier is configured so that a first word that is similar to another word in a specific one of the one or more phrases is not added to the specific phrase.
80. The system of claim 76 wherein the phrase identifier is configured so that a specific phrase is ended when a second word that is similar to another word already in the phrase is encountered.
81. The system of any one of claims 71 to 80 wherein the phrase identifier is configured so that noise words are ignored.
82. The system of any one of claims 71 to 80 wherein the phrase identifier is configured so that words that generally score highly in the corpus of documents are ignored.
83. The system of any one of claims 71 to 82 wherein the processor is configured to ignore those of the documents that are similar to other retrieved documents.
84. A computer-based system for identifying phrases related to an item from documents related to the item, the system comprising a computer, the computer comprising:
generator for generating a score for each of a plurality of words of the documents, the score indicating relatedness of a corresponding word to the item, wherein generating the score is based on one or more of:

average frequencies of the words in the documents;
average frequencies of the words in a corpus of documents;
and a fraction of the documents that contain the words; and a phrase identifier operable to, based on the scoring, identify one or more phrases relevant to the item from the documents wherein the identified phrases incorporate words in a top percentile of the scores.
85. The system of claim 84, further comprising a selector for selecting a description of the item and selecting, as related to the item, documents from the corpus of documents that match the selected description of the item.
86. The system of claim 85 wherein, for the selection of documents related to the item, the selector is configured to submit the selected description of the item to a search engine service and select documents based on search results provided by the search engine service.
87. The system of claim 85 wherein the selecting of the description includes the selector retrieving the description from an item catalog.
88. The system of claim 87 wherein the description is a title of the item that is stored in the item catalog.
89. The system of any one of claims 84 to 88 wherein the generator is configured to generate the score for a first word of the documents by:
calculating the average frequency of the first word by applying the formula:
where ~ is the average frequency of the first word, .function.i is the frequency of the first word in document i, and n is the number of the documents, and retrieving the normal frequency ~ of the first word, wherein the normal frequency represents the average frequency of the first word in the corpus of documents.
90. The system of claim 89, wherein the generator is configured to calculate a frequency score of the first word by applying the formula:
where S f is the frequency score of the first word.
91. The system of claim 90, wherein the generator is configured to calculate a contain score of the first word by applying the formula:
where S c is the contain score and n' is the number of documents that contain the first word.
92. The system of claim 91 wherein the generator is configured to generate the score of the first word by applying the formula:
S = .alpha. * S f + (1 ¨ .alpha.) * S c where S is the score of the first word and a is a constant between zero and one representing the weight given to the frequency score.
93. The system of claim 84, wherein the generator is configured to generate scores for phrases wherein a phrase is extended when extending would result in a score indicating that the extended phrase is more related than the unextended phrase to the item.
94. The system of claim 93 wherein a phrase is only extended by words with scores indicating a relatedness to the item.
95. The system of claim 94 wherein relatedness of a phrase to the item is determined to be high based on a number of occurrences of the phrase within the documents.
96. The system of any one of claims 93 to 95 wherein a word that is similar to another word in a phrase is not added to the phrase.
97. The system of any one of claims 93 to 96 wherein the phrase identifier is configured to end a phrase when a word that is similar to a word already in the phrase is encountered.
98. The system of any one of claims 84 to 98 wherein the phrase identifier is configured to ignore noise words.
99. The system of any one of claims 84 to 98 wherein the phrase identifier is configured to ignore words that generally score highly in a general corpus of documents.
100. The system of any one of claims 84 to 99 wherein the phrase identifier is configured to ignore documents that are similar to other ones of the documents.
101. The system of any one of claims 84 to 100 wherein the processor is configured to place an advertisement for the item with at least one search term that is the same as a phrase identified by the phrase identifier.
102. The system of any one of claims 84 to 101 wherein the processor is configured to automatically display an advertising message for the item to a user who has submitted a query containing a phrase among a set of phrases identified by the phrase identifier.
CA2575734A 2004-08-09 2005-08-08 Method and system for identifying keywords for use in placing keyword-targeted advertisements Expired - Fee Related CA2575734C (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US10/914,722 2004-08-09
US10/914,722 US7752200B2 (en) 2004-08-09 2004-08-09 Method and system for identifying keywords for use in placing keyword-targeted advertisements
PCT/US2005/028148 WO2006020576A2 (en) 2004-08-09 2005-08-08 Method and system for identifying keywords for use in placing keyword-targeted advertisements

Publications (2)

Publication Number Publication Date
CA2575734A1 CA2575734A1 (en) 2006-02-23
CA2575734C true CA2575734C (en) 2017-01-03

Family

ID=35908078

Family Applications (1)

Application Number Title Priority Date Filing Date
CA2575734A Expired - Fee Related CA2575734C (en) 2004-08-09 2005-08-08 Method and system for identifying keywords for use in placing keyword-targeted advertisements

Country Status (7)

Country Link
US (3) US7752200B2 (en)
EP (1) EP1776648A4 (en)
JP (1) JP4672726B2 (en)
KR (1) KR100904787B1 (en)
CN (2) CN102708115B (en)
CA (1) CA2575734C (en)
WO (1) WO2006020576A2 (en)

Families Citing this family (57)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8655727B2 (en) * 2003-12-30 2014-02-18 Amazon Technologies, Inc. Method and system for generating and placing keyword-targeted advertisements
US20050149388A1 (en) * 2003-12-30 2005-07-07 Scholl Nathaniel B. Method and system for placing advertisements based on selection of links that are not prominently displayed
US20070300142A1 (en) 2005-04-01 2007-12-27 King Martin T Contextual dynamic advertising based upon captured rendered text
US7752200B2 (en) 2004-08-09 2010-07-06 Amazon Technologies, Inc. Method and system for identifying keywords for use in placing keyword-targeted advertisements
GB2420426A (en) * 2004-11-17 2006-05-24 Transversal Corp Ltd An information handling system
EP1949273A1 (en) 2005-11-16 2008-07-30 Evri Inc. Extending keyword searching to syntactically and semantically annotated data
US8036937B2 (en) 2005-12-21 2011-10-11 Ebay Inc. Computer-implemented method and system for enabling the automated selection of keywords for rapid keyword portfolio expansion
US7752190B2 (en) 2005-12-21 2010-07-06 Ebay Inc. Computer-implemented method and system for managing keyword bidding prices
US7792858B2 (en) * 2005-12-21 2010-09-07 Ebay Inc. Computer-implemented method and system for combining keywords into logical clusters that share similar behavior with respect to a considered dimension
US9892425B2 (en) * 2006-03-27 2018-02-13 Genesys Telecommunications Laboratories, Inc. System for managing dynamic placement of advertisements in electronic information pages
US7908165B2 (en) * 2006-04-19 2011-03-15 Knee-Bone, Inc. Method and apparatus for generating performance curves
WO2007141020A1 (en) * 2006-06-06 2007-12-13 Exbiblio B.V. Contextual dynamic advertising based upon captured rendered text
US7890490B1 (en) * 2006-06-30 2011-02-15 United Video Properties, Inc. Systems and methods for providing advanced information searching in an interactive media guidance application
US20100057559A1 (en) * 2006-06-30 2010-03-04 Saar Wilf method of choosing advertisements to be shown to a search engine user
US20080027800A1 (en) * 2006-07-28 2008-01-31 Yves Schabes Method and apparatus for selecting related terms for electronic advertising
US20110066624A1 (en) * 2006-08-01 2011-03-17 Divyank Turakhia system and method of generating related words and word concepts
US7593935B2 (en) * 2006-10-19 2009-09-22 Paxfire Methods and systems for node ranking based on DNS session data
US7657514B2 (en) 2007-01-30 2010-02-02 Google Inc. Content identification expansion
US20080222133A1 (en) * 2007-03-08 2008-09-11 Anthony Au System that automatically identifies key words & key texts from a source document, such as a job description, and apply both (key words & text) as context in the automatic matching with another document, such as a resume, to produce a numerically scored result.
KR20090000611A (en) * 2007-03-08 2009-01-08 엔에이치엔(주) Advertisement method and system for displaying optimum title and description by analyzing click statistics
CA2717462C (en) 2007-03-14 2016-09-27 Evri Inc. Query templates and labeled search tip system, methods, and techniques
US7873640B2 (en) * 2007-03-27 2011-01-18 Adobe Systems Incorporated Semantic analysis documents to rank terms
US8788320B1 (en) 2007-03-28 2014-07-22 Amazon Technologies, Inc. Release advertisement system
KR100881832B1 (en) * 2007-03-30 2009-02-03 엔에이치엔(주) Method and system for displaying keyword advertisement using searching optimum randing page
KR100912168B1 (en) * 2007-06-08 2009-08-14 엔에이치엔비즈니스플랫폼 주식회사 Method and system of exposing advertisement automatically
KR100910521B1 (en) * 2007-08-14 2009-07-31 엔에이치엔비즈니스플랫폼 주식회사 Method for proposing advertisement keyword and system thereof
US20090076886A1 (en) * 2007-09-14 2009-03-19 Google Inc. Advertisement plusbox
US8594996B2 (en) 2007-10-17 2013-11-26 Evri Inc. NLP-based entity recognition and disambiguation
WO2009052308A1 (en) 2007-10-17 2009-04-23 Roseman Neil S Nlp-based content recommender
US8515791B2 (en) * 2007-11-02 2013-08-20 Buysafe, Inc. Method, system and components for obtaining, evaluating and/or utilizing seller, buyer and transaction data
US8744917B2 (en) * 2007-11-02 2014-06-03 Buysafe, Inc. Method, system and components for obtaining, evaluating and/or utilizing seller, buyer and transaction data
KR100980578B1 (en) * 2008-03-31 2010-09-06 엔에이치엔비즈니스플랫폼 주식회사 System and method for offering search result using registering extended keyword
US8145654B2 (en) * 2008-06-20 2012-03-27 Lexisnexis Group Systems and methods for document searching
US20100030648A1 (en) * 2008-08-01 2010-02-04 Microsoft Corporation Social media driven advertisement targeting
US9892103B2 (en) * 2008-08-18 2018-02-13 Microsoft Technology Licensing, Llc Social media guided authoring
US20100057772A1 (en) * 2008-08-29 2010-03-04 Microsoft Corporation Automatic determination of an entity's searchable social network using role-based inferences
EP2164247A3 (en) * 2008-09-12 2011-08-24 Axel Springer Digital TV Guide GmbH Method for distributing second multi-media content items in a list of first multi-media content items
US9521013B2 (en) * 2008-12-31 2016-12-13 Facebook, Inc. Tracking significant topics of discourse in forums
US8462160B2 (en) 2008-12-31 2013-06-11 Facebook, Inc. Displaying demographic information of members discussing topics in a forum
CA2796408A1 (en) * 2009-04-16 2010-10-21 Evri Inc. Enhanced advertisement targeting
US8266006B2 (en) 2009-11-03 2012-09-11 Ebay Inc. Method, medium, and system for keyword bidding in a market cooperative
US8554619B2 (en) * 2009-12-31 2013-10-08 Google Inc. Multi-campaign content allocation
US9710556B2 (en) 2010-03-01 2017-07-18 Vcvc Iii Llc Content recommendation based on collections of entities
US8788342B2 (en) * 2010-03-04 2014-07-22 Yahoo! Inc. Intelligent feature expansion of online text ads
US8645125B2 (en) 2010-03-30 2014-02-04 Evri, Inc. NLP-based systems and methods for providing quotations
US9405848B2 (en) 2010-09-15 2016-08-02 Vcvc Iii Llc Recommending mobile device activities
JP5535022B2 (en) * 2010-09-30 2014-07-02 ニフティ株式会社 Ad text creation support system
US8725739B2 (en) 2010-11-01 2014-05-13 Evri, Inc. Category-based content recommendation
US20120123863A1 (en) * 2010-11-13 2012-05-17 Rohit Kaul Keyword publication for use in online advertising
KR101722670B1 (en) 2010-12-28 2017-04-03 네이버 주식회사 System and method for recommending keyword based on indirect click
US8700487B2 (en) 2012-02-03 2014-04-15 Buysafe, Inc. User to website guaranteed shopping
US9792629B2 (en) * 2013-08-05 2017-10-17 Yahoo Holdings, Inc. Keyword recommendation
US20150170248A1 (en) * 2013-12-12 2015-06-18 Sears Brands, Llc Product and content association
CN106445963B (en) * 2015-08-10 2021-11-23 北京奇虎科技有限公司 Advertisement index keyword automatic generation method and device of APP platform
US10445376B2 (en) 2015-09-11 2019-10-15 Microsoft Technology Licensing, Llc Rewriting keyword information using search engine results
US20170116629A1 (en) * 2015-10-26 2017-04-27 International Business Machines Corporation System for searching existing customer experience information through cross-industries from text descriptions on a customer experience
US10515379B2 (en) * 2016-12-20 2019-12-24 Adobe Inc. Computerized detection and semantic characterization of trends in digital media content

Family Cites Families (101)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4992940A (en) 1989-03-13 1991-02-12 H-Renee, Incorporated System and method for automated selection of equipment for purchase through input of user desired specifications
US5099319A (en) 1989-10-23 1992-03-24 Esch Arthur G Video information delivery method and apparatus
US5319544A (en) 1989-11-20 1994-06-07 Itt Corporation Computerized inventory monitoring and verification system and method
EP0574529A1 (en) 1991-03-05 1993-12-22 The Gift Certificate Center, Inc. Method and apparatus for generating gift certificates
US5471382A (en) 1994-01-10 1995-11-28 Informed Access Systems, Inc. Medical network management system and process
US5598729A (en) 1994-10-26 1997-02-04 Tandem Systems, Inc. System and method for constructing wall of a tube
US5745602A (en) * 1995-05-01 1998-04-28 Xerox Corporation Automatic method of selecting multi-word key phrases from a document
US5615109A (en) 1995-05-24 1997-03-25 Eder; Jeff Method of and system for generating feasible, profit maximizing requisition sets
WO1997022066A1 (en) 1995-12-15 1997-06-19 The Softpages, Inc. Method for computer aided advertisement
JP3143079B2 (en) * 1997-05-30 2001-03-07 松下電器産業株式会社 Dictionary index creation device and document search device
US5937392A (en) * 1997-07-28 1999-08-10 Switchboard Incorporated Banner advertising display system and method with frequency of advertisement control
WO1999012108A1 (en) * 1997-09-04 1999-03-11 British Telecommunications Public Limited Company Methods and/or systems for selecting data sets
US5845278A (en) * 1997-09-12 1998-12-01 Inioseek Corporation Method for automatically selecting collections to search in full text searches
US6112021A (en) * 1997-12-19 2000-08-29 Mitsubishi Electric Information Technology Center America, Inc, (Ita) Markov model discriminator using negative examples
JP2000067081A (en) * 1998-08-24 2000-03-03 Matsushita Electric Ind Co Ltd Method for retrieving document and recording medium recorded with the same program and device for retrieving document
US6078866A (en) * 1998-09-14 2000-06-20 Searchup, Inc. Internet site searching and listing service based on monetary ranking of site listings
JP3389948B2 (en) * 1998-11-27 2003-03-24 日本電気株式会社 Display ad selection system
US6920464B2 (en) 1998-12-03 2005-07-19 Grantley Patent Holdings, Ltd. System for generating an advertising revenue projection
US6128651A (en) * 1999-04-14 2000-10-03 Americom Usa Internet advertising with controlled and timed display of ad content from centralized system controller
US7225182B2 (en) * 1999-05-28 2007-05-29 Overture Services, Inc. Recommending search terms using collaborative filtering and web spidering
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
US7035812B2 (en) * 1999-05-28 2006-04-25 Overture Services, Inc. System and method for enabling multi-element bidding for influencing a position on a search result list generated by a computer network search engine
US6449657B2 (en) 1999-08-06 2002-09-10 Namezero.Com, Inc. Internet hosting system
US8914361B2 (en) * 1999-09-22 2014-12-16 Google Inc. Methods and systems for determining a meaning of a document to match the document to content
US6963863B1 (en) * 1999-09-28 2005-11-08 Thomas Bannon Network query and matching system and method
US7392185B2 (en) * 1999-11-12 2008-06-24 Phoenix Solutions, Inc. Speech based learning/training system using semantic decoding
US7031932B1 (en) * 1999-11-22 2006-04-18 Aquantive, Inc. Dynamically optimizing the presentation of advertising messages
US20020010757A1 (en) 1999-12-03 2002-01-24 Joel Granik Method and apparatus for replacement of on-line advertisements
US6751621B1 (en) * 2000-01-27 2004-06-15 Manning & Napier Information Services, Llc. Construction of trainable semantic vectors and clustering, classification, and searching using trainable semantic vectors
US6606644B1 (en) 2000-02-24 2003-08-12 International Business Machines Corporation System and technique for dynamic information gathering and targeted advertising in a web based model using a live information selection and analysis tool
US7912868B2 (en) * 2000-05-02 2011-03-22 Textwise Llc Advertisement placement method and system using semantic analysis
US7076443B1 (en) * 2000-05-31 2006-07-11 International Business Machines Corporation System and technique for automatically associating related advertisements to individual search results items of a search result set
US7917390B2 (en) * 2000-06-09 2011-03-29 Sony Corporation System and method for providing customized advertisements over a network
CA2924940A1 (en) * 2000-07-05 2002-01-10 Paid Search Engine Tools, L.L.C. Paid search engine bid management
US6915294B1 (en) * 2000-08-18 2005-07-05 Firstrain, Inc. Method and apparatus for searching network resources
US7080073B1 (en) * 2000-08-18 2006-07-18 Firstrain, Inc. Method and apparatus for focused crawling
US6895406B2 (en) * 2000-08-25 2005-05-17 Seaseer R&D, Llc Dynamic personalization method of creating personalized user profiles for searching a database of information
US7689510B2 (en) 2000-09-07 2010-03-30 Sonic Solutions Methods and system for use in network management of content
US20020103698A1 (en) * 2000-10-31 2002-08-01 Christian Cantrell System and method for enabling user control of online advertising campaigns
DE10057634C2 (en) * 2000-11-21 2003-01-30 Bosch Gmbh Robert Process for processing text in a computer unit and computer unit
US20020111847A1 (en) 2000-12-08 2002-08-15 Word Of Net, Inc. System and method for calculating a marketing appearance frequency measurement
US7440943B2 (en) * 2000-12-22 2008-10-21 Xerox Corporation Recommender system and method
US20020094868A1 (en) * 2001-01-16 2002-07-18 Alma Tuck Methods for interactive internet advertising, apparatuses and systems including same
US7376588B1 (en) 2001-02-28 2008-05-20 Amazon.Com, Inc. Personalized promotion of new content
US6823333B2 (en) * 2001-03-02 2004-11-23 The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration System, method and apparatus for conducting a keyterm search
CN1227611C (en) * 2001-03-09 2005-11-16 北京大学 Method for judging position correlation of a group of query keys or words on network page
JP2002278949A (en) 2001-03-19 2002-09-27 Atr Onsei Gengo Tsushin Kenkyusho:Kk Device and method for generating title
US7058590B2 (en) 2001-05-04 2006-06-06 Hewlett-Packard Development Company, L.P. System and method for generating conversion-related estimates utilizing adaptive sample size
US7672894B2 (en) 2001-07-20 2010-03-02 Shopzilla, Inc. Automated bidding system for use with online auctions
US7076479B1 (en) 2001-08-03 2006-07-11 Overture Services, Inc. Search engine account monitoring
US6826572B2 (en) * 2001-11-13 2004-11-30 Overture Services, Inc. System and method allowing advertisers to manage search listings in a pay for placement search system using grouping
US20030120641A1 (en) * 2001-11-19 2003-06-26 Stephen Pelletier Method and apparatus for dynamic database creation and interactive analysis
US6763362B2 (en) * 2001-11-30 2004-07-13 Micron Technology, Inc. Method and system for updating a search engine
US7363254B2 (en) 2001-11-30 2008-04-22 Skinner Christopher J Automated web ranking bid management account system
US7295996B2 (en) * 2001-11-30 2007-11-13 Skinner Christopher J Automated web ranking bid management account system
US20030120626A1 (en) * 2001-12-20 2003-06-26 Koninklijke Philips Electronics N.V. Voice-enabled, consumer transaction system
US7346606B2 (en) 2003-06-30 2008-03-18 Google, Inc. Rendering advertisements with documents having one or more topics using user topic interest
US20050114198A1 (en) * 2003-11-24 2005-05-26 Ross Koningstein Using concepts for ad targeting
DE10216332A1 (en) 2002-04-13 2003-10-30 Conducta Endress & Hauser Measuring device for process technology and operating procedures for a measuring device
US20030208394A1 (en) 2002-05-01 2003-11-06 Todd Burris Sales tracking and forecasting application tool
US20030216930A1 (en) 2002-05-16 2003-11-20 Dunham Carl A. Cost-per-action search engine system, method and apparatus
US20040044574A1 (en) 2002-06-04 2004-03-04 Kordex, Inc. Apparatus for displaying local advertising to a display screen
US20040044571A1 (en) * 2002-08-27 2004-03-04 Bronnimann Eric Robert Method and system for providing advertising listing variance in distribution feeds over the internet to maximize revenue to the advertising distributor
US7904453B2 (en) * 2002-10-17 2011-03-08 Poltorak Alexander I Apparatus and method for analyzing patent claim validity
US7158957B2 (en) * 2002-11-21 2007-01-02 Honeywell International Inc. Supervised self organizing maps with fuzzy error correction
WO2004051418A2 (en) * 2002-11-27 2004-06-17 Findwhat.Com System and method for pay for performance advertising having biddable advertising units utilizing rotating routing to advertiser websites
JP3944102B2 (en) * 2003-03-13 2007-07-11 株式会社日立製作所 Document retrieval system using semantic network
JP2004280661A (en) * 2003-03-18 2004-10-07 Fujitsu Ltd Retrieval method and program
US7127408B2 (en) 2003-06-13 2006-10-24 Rosen Michael J Method of creating season ticket package
US7363302B2 (en) * 2003-06-30 2008-04-22 Googole, Inc. Promoting and/or demoting an advertisement from an advertising spot of one type to an advertising spot of another type
US8438154B2 (en) 2003-06-30 2013-05-07 Google Inc. Generating information for online advertisements from internet data and traditional media data
EP1661008A4 (en) * 2003-08-05 2007-01-24 Cnet Networks Inc Product placement engine and method
US11042886B2 (en) * 2003-09-04 2021-06-22 Google Llc Systems and methods for determining user actions
US7647242B2 (en) * 2003-09-30 2010-01-12 Google, Inc. Increasing a number of relevant advertisements using a relaxed match
US7346839B2 (en) 2003-09-30 2008-03-18 Google Inc. Information retrieval based on historical data
US20050097024A1 (en) 2003-10-30 2005-05-05 Rainey Jim E. Multi-party bidding for online advertising space
US20040133469A1 (en) * 2003-11-04 2004-07-08 Dario Chang System and method of promote website using Cycle Hits and Hits History
US20050144158A1 (en) 2003-11-18 2005-06-30 Capper Liesl J. Computer network search engine
US7634472B2 (en) 2003-12-01 2009-12-15 Yahoo! Inc. Click-through re-ranking of images and other data
US20050137939A1 (en) * 2003-12-19 2005-06-23 Palo Alto Research Center Incorporated Server-based keyword advertisement management
US20050144064A1 (en) 2003-12-19 2005-06-30 Palo Alto Research Center Incorporated Keyword advertisement management
US20050144068A1 (en) * 2003-12-19 2005-06-30 Palo Alto Research Center Incorporated Secondary market for keyword advertising
US20050144069A1 (en) * 2003-12-23 2005-06-30 Wiseman Leora R. Method and system for providing targeted graphical advertisements
US8655727B2 (en) * 2003-12-30 2014-02-18 Amazon Technologies, Inc. Method and system for generating and placing keyword-targeted advertisements
CA2550853C (en) 2003-12-30 2020-03-31 Amazon Technologies, Inc. Method and system for placing advertisements based on selection of links that are not prominently displayed
US20050149388A1 (en) 2003-12-30 2005-07-07 Scholl Nathaniel B. Method and system for placing advertisements based on selection of links that are not prominently displayed
US8392249B2 (en) * 2003-12-31 2013-03-05 Google Inc. Suggesting and/or providing targeting criteria for advertisements
US7424421B2 (en) * 2004-03-03 2008-09-09 Microsoft Corporation Word collection method and system for use in word-breaking
US20050216335A1 (en) * 2004-03-24 2005-09-29 Andrew Fikes System and method for providing on-line user-assisted Web-based advertising
US8510161B2 (en) 2004-05-03 2013-08-13 Target Brands, Inc. Reservation gift card
US20050267872A1 (en) * 2004-06-01 2005-12-01 Yaron Galai System and method for automated mapping of items to documents
US8494900B2 (en) 2004-06-30 2013-07-23 Google Inc. Adjusting ad costs using document performance or document collection performance
US7752200B2 (en) 2004-08-09 2010-07-06 Amazon Technologies, Inc. Method and system for identifying keywords for use in placing keyword-targeted advertisements
US7949561B2 (en) * 2004-08-20 2011-05-24 Marketing Evolution Method for determining advertising effectiveness
WO2006035196A1 (en) * 2004-09-30 2006-04-06 British Telecommunications Public Limited Company Information retrieval
US7467349B1 (en) 2004-12-15 2008-12-16 Amazon Technologies, Inc. Method and system for displaying a hyperlink at multiple levels of prominence based on user interaction
US8429184B2 (en) * 2005-12-05 2013-04-23 Collarity Inc. Generation of refinement terms for search queries
US7752190B2 (en) 2005-12-21 2010-07-06 Ebay Inc. Computer-implemented method and system for managing keyword bidding prices
US7873640B2 (en) * 2007-03-27 2011-01-18 Adobe Systems Incorporated Semantic analysis documents to rank terms
US8788320B1 (en) 2007-03-28 2014-07-22 Amazon Technologies, Inc. Release advertisement system
US8335719B1 (en) 2007-06-26 2012-12-18 Amazon Technologies, Inc. Generating advertisement sets based on keywords extracted from data feeds

Also Published As

Publication number Publication date
CN102708115A (en) 2012-10-03
US20060041536A1 (en) 2006-02-23
US7752200B2 (en) 2010-07-06
CA2575734A1 (en) 2006-02-23
CN102708115B (en) 2015-12-09
WO2006020576A2 (en) 2006-02-23
WO2006020576A3 (en) 2007-01-25
JP2008509501A (en) 2008-03-27
EP1776648A4 (en) 2011-01-26
KR20070064320A (en) 2007-06-20
US10402431B2 (en) 2019-09-03
CN101023408A (en) 2007-08-22
KR100904787B1 (en) 2009-06-25
US20170103122A1 (en) 2017-04-13
EP1776648A2 (en) 2007-04-25
JP4672726B2 (en) 2011-04-20
US9489449B1 (en) 2016-11-08

Similar Documents

Publication Publication Date Title
CA2575734C (en) Method and system for identifying keywords for use in placing keyword-targeted advertisements
US10176495B2 (en) Selection of keyword phrases for providing contextually relevant content to users
US9916366B1 (en) Query augmentation
JP4647623B2 (en) Universal search engine interface
US8321278B2 (en) Targeted advertisements based on user profiles and page profile
US8370343B2 (en) Selecting directly bid upon advertisements for display
KR100857049B1 (en) Automatically targeting web-based advertisements
US20090063265A1 (en) Information network for text ads
US20050149390A1 (en) Method and system for generating and placing keyword-targeted advertisements
EP1591923A1 (en) Method and system for ranking documents of a search result to improve diversity and information richness
US20010049677A1 (en) Methods and systems for enabling efficient retrieval of documents from a document archive
WO2009006150A1 (en) Using previous user search query to target advertisements
KR20070038146A (en) Personalization of placed content ordering in search results
KR20060036065A (en) Serving advertisements using a search of advertiser web information
TW200842632A (en) Related search queries for a webpage and their applications
US20080243826A1 (en) System and method for determining semantically related terms
US20050149388A1 (en) Method and system for placing advertisements based on selection of links that are not prominently displayed
US8224693B2 (en) Advertisement selection based on key words
US20120185457A1 (en) Creating business value by embedding domain tuned search on web-sites
CA2550853C (en) Method and system for placing advertisements based on selection of links that are not prominently displayed
Craswell et al. Web information retrieval

Legal Events

Date Code Title Description
EEER Examination request
MKLA Lapsed

Effective date: 20220301

MKLA Lapsed

Effective date: 20200831