Systems and Methods for Finding Keyword Relationships Using Wisdoms from Multiple Sources

• US 20110208708A1
• Filed: 02/25/2010
• Published: 08/25/2011
• Est. Priority Date: 02/25/2010
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Zhang, Qiong, Liu, Weiguo
• Granted Patent: US 8,214,348 B2
• Time in Patent Office: Days
• US Class Current: 707/706
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06Q 30/0241 Advertisements

Abstract

Systems and methods for finding related terms based on three different sources are disclosed. Generally, a first plurality of distances is determined based on one or more received terms and a first plurality of terms derived from an algorithmic search list. A second plurality of distances is determined based on the one or more received terms and a second plurality of terms derived from a sponsored search list. A third plurality of distances is determined based on the one or more received terms and a third plurality of terms derived from search logs. The first, second, and third pluralities of distances are combined to derive a fourth plurality of distances. Finally, a plurality of related terms related to the one or more received terms is generated based on the fourth plurality of distances.

Claims

1. A computer implemented method for determining related terms, the method comprising:

receiving one or more terms;

determining, in a cloud computing environment, a first plurality of distances based on the one or more received terms and a first plurality of terms that is derived from an algorithmic search list, wherein a search engine server generates the first plurality of terms based on the one or more received terms;

determining, in the cloud computing environment, a second plurality of distances based on the one or more received terms and a second plurality of terms that is derived from a sponsored search list, wherein the second plurality of terms is generated based on the one or more received terms;

determining, in the cloud computing environment, a third plurality of distances based on the one or more received terms and a third plurality of terms, wherein the third plurality of terms is derived from a search log;

combining the first, second, and third pluralities of distances to derive a fourth plurality of distances; and

generating a plurality of related terms related to the one or more received terms based on the fourth plurality of distances.

2. The computer implemented method of claim 1, wherein the one or more received terms comprises at least one of the following:

a seed term, a keyword, a function of keywords, and an instruction to choose keywords.

3. The computer implemented method of claim 1, wherein a sponsored search server generates the sponsored search list based on the one or more received terms.

4. The computer implemented method of claim 1, wherein generating a plurality of related terms related to the one or more received terms based on the fourth plurality of distances comprises:

recommending the plurality of related terms related to the one or more received terms when the corresponding combined distance is less than a predetermined value.

5. The computer implemented method of claim 1, wherein the related terms comprises at least one of text, voice, image, and video.

6. The computer implemented method of claim 1, wherein determining the first plurality of distances between the one or more received terms and the first plurality of terms from the search engine server comprises:

weighing the one or more received terms and the first plurality of terms by term frequency inverse document frequency; and

calculating a plurality of Euclidian distances between the one or more received terms and the first plurality of terms.

7. The computer implemented method of claim 1, wherein determining the first plurality of distances between the one or more received terms and the first plurality of terms from the search engine server comprises:

applying probabilistic latent semantic indexing (PLSI) to the one or more received terms and a word bag comprising the first plurality of terms; and

calculating a plurality of Kullback-Leibler (KL) divergences between the one or more received terms and the first plurality of terms in the word bag.

8. The computer implemented method of claim 1, wherein determining the second plurality of distances between the one or more received terms and the second plurality of terms from the advertisement server comprises:

weighing the one or more received terms and the second plurality of terms by term frequency inverse document frequency; and

calculating a plurality of Euclidian distances between the one or more received terms and the second plurality of terms.

9. The computer implemented method of claim 1, wherein the search log comprises Uniform Resource Locators (URLs) from search queries and search clicks in a search session.

10. The computer implemented method of claim 1, wherein determining the second plurality of distances between the one or more received terms and the second plurality of terms from the advertisement server comprises:

applying probabilistic latent semantic indexing (PLSI) to the one or more received terms and a word bag comprising the second plurality of terms; and

calculating a plurality of Kullback-Leibler (KL) divergences between the one or more received terms and the second plurality of terms in the word bag.

11. The computer implemented method of claim 9, wherein a total duration of the search session is less than a first predetermined threshold.

12. The computer implemented method of claim 9, wherein a longest duration of inactivity in the search session is less than a second predetermined threshold.

13. The computer implemented method of claim 9, wherein determining the third plurality of distances between the one or more received terms and the third plurality of terms from the search log comprises:

measuring a first plurality of similarities between previous one or more received terms and terms from the search log;

measuring a second plurality of similarities between a pair of previous one or more received terms and a URL;

measuring a third plurality of similarities between a pair of URLs; and

leveraging the first, second, and third pluralities of similarities to determine the third plurality of distances.

14. The computer implemented method of claim 9, wherein combining the first, second, and third pluralities of distances to the fourth plurality of distances comprises:

combining linearly a pair of distances from the first, second, and third pluralities of distances.

15. The computer implemented method of claim 9, wherein combining the first, second, and third pluralities of distances to the fourth plurality of distances comprises:

multiplying a pair of distances from the first, second, and third pluralities of distances.

29. The computer implemented method of claim 1, wherein the one or more received terms comprises at least one of the following:

a term from a user terminal and a term from a data file.

16. A computer-readable storage medium storing a set of instructions for generating related terms based on one or more terms, the set of instructions to direct a processor to perform acts of:

receiving one or more terms;

determining a first plurality of distances based on the one or more received terms and a first plurality of terms that is derived from an algorithmic search list generated by a search engine server based on the one or more received terms;

determining a second plurality of distances based on the one or more received terms and a second plurality of terms that is derived from a sponsored search list generated based on the one or more received terms;

determining a third plurality of distances based on the one or more received terms and a third plurality of terms that is derived from a search log;

combining the first, second, and third pluralities of distances to derive a fourth plurality of distances; and

generating a plurality of related terms related to the one or more received terms based on the fourth plurality of distances.

17. The computer-readable storage medium of claim 16, wherein determining the first plurality of distances between the one or more received terms and the first plurality of terms from the search engine server comprises:

weighing the one or more received terms and the first plurality of terms by term frequency inverse document frequency; and

calculating a plurality of Euclidian distances between the one or more received terms and the first plurality of terms.

18. The computer-readable storage medium of claim 16, wherein determining the first plurality of distances between the one or more received terms and the first plurality of terms from the search engine server comprises:

applying probabilistic latent semantic indexing (PLSI) to the one or more received terms and a word bag comprising the first plurality of terms; and

calculating a plurality of Kullback-Leibler (KL) divergences between the one or more received terms and the first plurality of terms in the word bag.

19. The computer-readable storage medium of claim 16, wherein determining the second plurality of distances between the one or more received terms and the second plurality of terms from the advertisement server comprises:

applying probabilistic latent semantic indexing (PLSI) to the one or more received terms and a word bag comprising the second plurality of terms; and

calculating a plurality of Kullback-Leibler (KL) divergences between the one or more received terms and the second plurality of terms in the word bag.

20. The computer-readable storage medium of claim 16, wherein the search log comprises Uniform Resource Locators (URLs) from search queries and search clicks in a search session.

21. The computer implemented method of claim 20, wherein a total duration of the search session is less than a first predetermined threshold.

22. The computer implemented method of claim 20, wherein a longest duration of inactivity in the search session is less than a second predetermined threshold.

23. A computer system for generating related terms based on an one or more terms, the computer system comprising:

a first module configured to receive one or more terms;

a second module configured to determine a first plurality of distances based on the one or more received terms and a first plurality of terms that is derived from an algorithmic search list generated by a search engine server based on the one or more received terms;

a third module configured to determine a second plurality of distances based on the one or more received terms and a second plurality of terms that is derived from a sponsored search list generated based on the one or more received terms;

a fourth module configured to determine a third plurality of distances based on the one or more received terms and a third plurality of terms that is derived from a search log;

a fifth module configured to combine the first, second, and third pluralities of distances to derive a fourth plurality of distances; and

a sixth module configured to generate a plurality of related terms related to the one or more received terms based on the fourth plurality of distances.

wherein the search log comprises Uniform Resource Locators (URLs) from search queries and search clicks in a search session.

24. The computer system of claim 23, wherein a total duration of the search session is less than a first predetermined threshold.

25. The computer system of claim 23, wherein a longest duration of inactivity in the search session is less than a second predetermined threshold.

26. The computer system of claim 23, wherein the sponsored search list is generated by the search engine server based on the one or more received terms.

27. The computer system of claim 23, wherein the sixth module recommends the plurality of related terms related to the one or more received terms when the corresponding combined distance is less than a predetermined value.

28. The computer system of claim 23, wherein the first module is configured to receive one or more terms from at least one of the following:

a user terminal and a data file.