Inverse search systems and methods

• US 20050216454A1
• Filed: 03/15/2005
• Published: 09/29/2005
• Est. Priority Date: 03/15/2004
• Status: Active Grant

• Current Assignee: Pinterest, Inc.
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Nguyen, Nam, Lee, Kevin, Walther, Eckart, Lu, Qi, Diab, Ali, Ku, David
• Granted Patent: US 8,150,825 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/245 Query processing
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

Inverse search systems and methods operate on identifiers of content items in a corpus such as the World Wide Web In an inverse search, the user submits a query that includes an identifier of a target content item in the corpus and receives information (metadata) about the target content item being returned to the user. Many types of metadata can be returned, including ratings or other metadata related to the target content item obtained from users, popularity data specific to the target content item, information about previously submitted forward search queries that led to the target content item being identified as a hit, and metadata extracted from the target content item.

Claims

1. In a computer system, a method for performing an inverse search, the method comprising:

receiving an inverse search query from a querying user, the inverse search query including a target content identifier;

obtaining metadata associated with the target content identifier from a data store containing metadata associated with particular content identifiers, wherein at least some of the metadata was collected during previous interactions of users with the computer system;

generating an inverse search report based at least in part on the obtained metadata; and

transmitting the inverse search report to the querying user.

2. The method of claim 1, wherein the inverse search report includes a popularity score for the target content identifier.

3. The method of claim 2 wherein the act of obtaining the metadata includes retrieving the popularity score from the data store.

4. The method of claim 2 wherein the act of obtaining the metadata includes:

retrieving raw popularity data from the data store; and

computing the popularity score from the raw popularity data.

5. The method of claim 2 wherein the computer system includes a forward search module and wherein the popularity score is based on a clickthrough rate for the target content identifier measured from previous forward searches executed by the computer system.

6. The method of claim 5 wherein the clickthrough rate is defined as N_click/N_hit, where N_hit is a number of previous instances of forward searches executed by the computer system for which the target content identifier was returned as a hit and N_click is the number of such instances that resulted in user navigation to the content identified by the target content identifier.

7. The method of claim 1 wherein the computer system includes a forward search module and wherein the act of obtaining the metadata includes:

retrieving from the data store a plurality of referral queries for the target content identifier, wherein each referral query is a forward search query previously submitted to the computer system for which the computer system returned the target content identifier as a hit,

wherein the plurality of referral queries is included in the inverse search report.

8. The method of claim 7 wherein the referral queries are included in the inverse search report in a list order based at least in part on a clickthrough rate for the target content identifier relative to each of the referral queries.

9. The method of claim 1 wherein the act of obtaining the metadata includes:

retrieving from the data store user-specific metadata associated with the target content identifier, wherein the user-specific metadata was previously received from the querying user,

wherein the user-specific metadata is included in the inverse search report.

10. The method of claim 9 wherein the user-specific metadata includes a user-supplied rating of a target content item identified by the target content identifier.

11. The method of claim 1 wherein the act of obtaining the metadata includes:

retrieving from the data store user-specific metadata associated with the target content identifier, wherein the user-specific metadata was previously received from a user other than the querying user,

wherein the user-specific metadata is included in the inverse search report.

12. The method of claim 11 wherein the user other than the querying user is a member of a trust network defined for the querying user.

13. The method of claim 11 wherein the user-specific metadata includes a user-supplied rating of a target content item identified by the target content identifier.

14. The method of claim 1 wherein the act of obtaining the metadata includes:

retrieving from the data store a plurality of annotations associated with the target content identifier, wherein each annotation was previously received from one of a plurality of users and wherein each annotation includes user-specific metadata; and

aggregating the user-specific metadata across the plurality of retrieved annotations,

wherein the aggregated metadata is included in the inverse search report.

15. The method of claim 14 wherein the plurality of users includes users who are members of a trust network defined for the querying user.

16. The method of claim 14 wherein the plurality of users includes all users of the computer system.

17. The method of claim 14 wherein the user-specific metadata of each annotation includes a user-supplied rating of a target content item identified by the target content identifier.

18. The method of claim 17 wherein the aggregated metadata includes an average rating of the target content item determined from the user-supplied ratings.

19. The method of claim 1 wherein the act of obtaining the metadata includes:

retrieving from the data store a global rating associated with the content identifier, wherein the global rating is computed from individual ratings of a target content item identified by the content identifier, each individual rating being submitted by a user of the search system,

wherein the global rating is included in the inverse search report.

20. The method of claim 1 wherein the inverse search report further includes metadata extracted from a target content item identified by the target content identifier.

21. The method of claim 20 wherein the metadata is extracted from the target content item in response to the inverse search query.

22. The method of claim 20 wherein the metadata is extracted from the target content item and stored in the data store prior to receiving the inverse search query, and wherein the act of obtaining the metadata includes retrieving the extracted metadata from the data store.

23. The method of claim 20 wherein the metadata extracted from the target content item includes a physical location of an owner or provider of the target content item.

24. The method of claim 20 wherein the metadata extracted from the target content item includes an indicator as to whether the target content item includes adult content.

25. The method of claim 20 wherein the metadata extracted from the target content item includes an indicator as to whether the target content item includes a pop-up window.

26. The method of claim 1 wherein the content identifier comprises a URL identifying a World Wide Web page or site.

27. The method of claim 1 wherein the querying user is a human.

28. The method of claim 1 wherein the querying user is a computer.

29. The method of claim 1 wherein the querying user communicates with the computer system via a network.

30. A computer system for performing an inverse search, the computer system comprising:

a data store configured to store metadata associated with particular content items;

a metadata collection module communicably coupled to the data store and configured such that, during a user interaction with the computer system, the metadata collection module adds to the data store metadata associated with particular content items; and

an inverse search server communicably coupled to the data store, the inverse search server including;

input control logic configured to receive an inverse search query from a querying user, the inverse search query including a target content identifier;

search control logic configured to obtain metadata associated with the target content identifier from the data store; and

reporting control logic configured to generate an inverse search report based at least in part on the metadata retrieved by the search control logic and further configured to transmit the inverse search report to the querying user.