System and methods for ranking the relative value of terms in a multi-term search query using deletion prediction

• US 7,406,465 B2
• Filed: 12/14/2004
• Issued: 07/29/2008
• Est. Priority Date: 12/14/2004
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Fain, Daniel C., Jones, Rosemary
• Primary Examiner(s): Cottingham; John
• Assistant Examiner(s): Rayyan; Susan F
• Application Number:
  US11/013,298
• Publication Number:
  US 20060129534A1
• Time in Patent Office: 1,323 Days
• Field of Search: 707/1, 707 3- 5, 707/6, 707/10, 707/102, 707/104.1, 705/26
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

The likely relevance of each term of a search-engine query of two or more terms is determined by their deletion probability scores. If the deletion probability scores are significantly different, the deletion probability score can be used to return targeted ads related to the more relevant term or terms along with the search results. Deletion probability scores are determined by first gathering historical records of search queries of two or more terms in which a subsequent query was submitted by the same user after one or more of the terms had been deleted. The deletion probability score for a particular term of a search query is calculated as the ratio of the number of times that particular term was itself deleted prior to a subsequent search by the same user divided by the number of times there were subsequent search queries by the same user in which any term or terms including that given term was deleted by the same user prior to the subsequent search. Terms are not limited to individual alphabetic words.

Claims

1. A computer-implemented method for identifying a relevant term in a subsequent multi-term search query, the method comprising:

tracking a number of deletions of each particular term in a plurality of multi-term search queries, wherein each of the plurality of multi-term search queries includes each particular term;

tracking a total number of deletions in the plurality of multi-term search queries;

determining deletion probabilities of each particular term from the number of deletions of each particular term and the total number of deletions;

determining a relevance value of the subsequent multi-term search query based upon the deletion probabilities of each particular term of the subsequent multi-term search query;

determining whether the relevance value exceeds a threshold value;

upon determining that the relevance value exceeds the threshold value, identifying the relevant term in the subsequent multi-term search query, the relevant term having a deletion probability less than deletion probabilities of remaining terms from the subsequent multi-term search query;

utilizing the relevant term to retrieve, from a recommended content store, a plurality of recommended content related to the relevant term, wherein the plurality of recommended content comprises at least one of an advertising link and a search result; and

displaying the plurality of recommended content to a user in a predetermined area on a client browser.

3. The computer implemented method of claim 1, wherein determining deletion probabilities of each particular term from the number of deletions of each particular term and the total number of deletions comprises dividing the number of deletions of a particular term by the total number of deletions in a selected one of the plurality of multi-term search queries that includes the particular term.

4. The computer implemented method of claim 3, comprising adding the deletion probability of each particular term to a list of deletion probability scores.

5. The computer implemented method of claim 4, wherein determining a relevance value of the subsequent multi-term search query based upon the deletion probabilities of each particular term of the subsequent multi-term search query comprises retrieving the deletion probability of each particular term of the subsequent multi-term search query from the list of deletion probability scores.

6. The computer implemented method of claim 1, wherein determining a relevance value of the subsequent multi-term search query based upon the deletion probabilities of each particular term of the subsequent multi-term search query comprises determining an absolute difference between the deletion probabilities of each particular term of the subsequent multi-term search query.

7. The computer implemented method of claim 6, wherein determining the absolute difference between the deletion probabilities of each particular term of the subsequent multi-term search query further comprises applying an adjustment factor.

2. A computer-implemented system for identifying a relevant term in a subsequent multi-term search query, the system comprising:

a recommended content store for storing content related to keywords;

a deletion predictor configured to track a number of deletions of each particular term in a plurality of multi-term search queries, wherein each of the plurality of multi-term search queries includes each particular term, and to track a total number of deletions in the plurality of multi-term search queries, and to determine deletion probabilities of each particular term from the number of deletions of each particular term and the total number of deletions;

a relevance identifier configured to determine a relevance value of the subsequent multi-term search query based upon the deletion probabilities of each particular term of the subsequent multi-term search query, determine whether the relevance value exceeds a threshold value; and

upon determining that the relevance value exceeds the threshold value, to identify the relevant term in the subsequent multi-term search query, the relevant term having a deletion probability less than deletion probabilities of remaining terms from the subsequent multi-term search query; and

a content server for receiving the relevant term and selecting a plurality of recommended content from the recommended content store, the selecting performed based on the relevant term and the plurality of recommended content being related to the relevant term, wherein the plurality of recommended content comprises at least one of an advertising link and a search result; and

a display for displaying the plurality of recommended content to a user in a predetermined area on a client browser.

8. The computer implemented system of claim 2, wherein the deletion predictor is configured to determine deletion probabilities of each particular term from the number of deletions of each particular term and the total number of deletions by dividing the number of deletions of a particular term by the total number of deletions in a selected one of the plurality of multi-term search queries that includes the particular term.

9. The computer implemented system of claim 8, comprising a deletion probability store for storing a list of deletion probability scores, wherein each deletion probability score is the deletion probability of each particular term.

10. The computer implemented system of claim 9, wherein the relevance identifier is configured to determine the relevance value of the subsequent multi-term search query based upon the deletion probabilities of each particular term of the subsequent multi-term search query by retrieving the deletion probability of each particular term of the subsequent multi-term search query from the list of deletion probability scores stored in the deletion probability store.

11. The computer implemented system of claim 2, wherein the relevance identifier is configured to determine the relevance value of the subsequent multi-term search query based upon the deletion probabilities of each particular term of the subsequent multi-term search query by determining an absolute difference between the deletion probabilities of each particular term of the subsequent multi-term search query.

12. The computer implemented method of claim 11, wherein the relevance identifier configured to determine the absolute difference between the deletion probabilities of each particular term of the subsequent multi-term search query is further configured to apply an adjustment factor.