Techniques for searching future events

• US 7,668,813 B2
• Filed: 08/11/2006
• Issued: 02/23/2010
• Est. Priority Date: 08/11/2006
• Status: Active Grant

• Current Assignee: University of Chile, Verizon Media, Inc. (Verizon Communications Inc.), Yahoo Assets LLC
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Baeza-Yates, Ricardo Alberto
• Primary Examiner(s): Ali; Mohammad
• Assistant Examiner(s): Shmatov; Alexey
• Application Number:
  US11/463,978
• Publication Number:
  US 20080040321A1
• Time in Patent Office: 1,292 Days
• Field of Search: 707/3
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2477 Temporal data queries
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

A future retrieval system operable to enable searching for future events associated with a set of documents includes at least one memory device configured to store a database of the documents. The system identifies future events associated with the documents and generates temporal data and a confidence value for each future event. The confidence value represents a likelihood of occurrence for the corresponding future event. The system evaluates text associated with the future events to generate event metadata, and indexes the documents with the associated event metadata, temporal data, and confidence levels in the database. The system generates search results corresponding to selected ones of the documents in the database in response to queries which include references to future times, and ranks the search results in accordance with the confidence values associated with the future events associated with the selected documents.

Claims

1. A future retrieval system operable to enable searching for future events associated with a set of documents, comprising:

at least one memory device configured to store a database which includes the documents; and

at least one computing device configured to;

identify one or more future events associated with each of the documents, at least some of the documents having a plurality of the future events associated therewith;

generate temporal data and a confidence value for each future event, the temporal data for each future event comprising a time segment, the confidence value representing a likelihood of occurrence for the corresponding future event and being determined with reference to text associated with the corresponding future event by identifying one or both of a verb tense or verb mood for the associated text, each future event corresponding to a tuple of the corresponding time segment and confidence value;

identify a text scope for at least some of the future events, the text scope corresponding to a subset of the document with which the corresponding future event is associated;

evaluate the text associated with the future events to generate event metadata for each future event;

index the subsets of the documents corresponding to each future event with the corresponding event metadata, and the corresponding tuple of the corresponding time segment and confidence value in the database; and

generate search results corresponding to selected ones of the documents in the database in response to queries which include references to future times, and to rank the search results in accordance with the confidence values associated with the future events associated with the selected documents.

2. The future retrieval system of claim 1 wherein the at least one computing device is configured to use a natural language processing technique to identify the verb mood for each of the future events.

3. The future retrieval system of claim 1 wherein the at least one computing device is configured to identify one or both of the verb tense or verb mood by identifying one or more of will, must, should, could, or would in the text.

4. The future retrieval system of claim 1 wherein the temporal data comprises one or more of a second, a minute, an hour, a day, a week, a month, a year, a decade, a century, or a millennium.

5. The future retrieval system of claim 1 wherein selected ones of the queries include one or more keywords, the at least one computing device further being configured to rank the search results in accordance with relevance of the selected documents with respect to the one or more keywords.

6. The future retrieval system of claim 1 wherein selected ones of the queries include only the references to the future times, and the selected documents corresponding to the search results include any of the documents referring to the future times.

7. The future retrieval system of claim 1 wherein first ones of the documents each have a plurality of the future events associated therewith, the at least one computing device further being configured to generate rank values for each of the first documents with reference to a most likely to occur one of the plurality of associated future events.

8. The future retrieval system of claim 1 wherein the at least one computing device is further configured to generate a ranking value for a first one of the selected documents in response to a first query identifying a time segment, the ranking value being determined with reference to a maximum confidence value associated with the first selected document restricted to the time segment.

9. The future retrieval system of claim 1 wherein a first one of the selected documents has a plurality of the future events associated therewith, and wherein the at least one computing device is further configured to generate a ranking value for the first selected document in response to a first query identifying a time segment, the ranking value being determined with reference to an intersection of the time segment with a curve representing the plurality of future events associated with the first selected document.

10. The future retrieval system of claim 1 wherein a first one of the documents has a plurality of the future events associated therewith, and wherein the at least one computing device is further configured to divide the first document into a plurality of text passages, each of the text passages being associated with only one of the plurality of future events associated with the first document.

11. A computer-implemented method for enabling searching for future events associated with a set of documents stored in a database on at least one memory device, the method comprising:

identifying one or more future events associated with each of the documents using one or more computing devices, at least some of the documents having a plurality of the future events associated therewith;

generating temporal data and a confidence value for each future event using the one or more computing devices, the temporal data for each future event comprising a time segment, the confidence value representing a likelihood of occurrence for the corresponding future event and being determined with reference to text associated with the corresponding future event by identifying one or both of a verb tense or verb mood for the associated text, each future event corresponding to a tuple of the corresponding time segment and confidence value;

identifying a text scope for at least some of the future events using the one or more computing devices, the text scope corresponding to a subset of the document with which the corresponding future event is associated;

evaluating the text associated with the future events to generate event metadata for each future event using the one or more computing devices;

indexing the subsets of the documents corresponding to each future event with the corresponding event metadata, and the corresponding tuple of the corresponding time segment and confidence value in the database; and

generating search results corresponding to selected ones of the documents in the database in response to queries which include references to future times, and ranking the search results in accordance with the confidence values associated with the future events associated with the selected documents.

12. The computer-implemented method of claim 11 wherein identifying one or both of the verb tense or verb mood comprises identifying one or more of will, must, should, could, or would in the text.

13. The computer-implemented method of claim 11 wherein the temporal data comprises one or more of a second, a minute, an hour, a day, a week, a month, a year, a decade, a century, or a millennium.

14. The computer-implemented method of claim 11 wherein selected ones of the queries include one or more keywords, the method further comprising ranking the search results in accordance with relevance of the selected documents with respect to the one or more keywords.

15. The computer-implemented method of claim 11 wherein selected ones of the queries include only the references to the future times, and the selected documents corresponding to the search results include any of the documents referring to the future times.

16. The computer-implemented method of claim 11 wherein first ones of the documents each have a plurality of the future events associated therewith, the method further comprising generating rank values for each of the first documents with reference to a most likely to occur one of the plurality of associated future events.

17. The computer-implemented method of claim 11 further comprising generating a ranking value for a first one of the selected documents in response to a first query identifying a time segment, the ranking value being determined with reference to a maximum confidence value associated with the first selected document restricted to the time segment.

18. The computer-implemented method of claim 11 wherein a first one of the selected documents has a plurality of the future events associated therewith, the method further comprising generating a ranking value for the first selected document in response to a first query identifying a time segment, the ranking value being determined with reference to an intersection of the time segment with a curve representing the plurality of future events associated with the first selected document.

19. The computer-implemented method of claim 11 wherein a first one of the documents has a plurality of the future events associated therewith, the method further comprising dividing the first document into a plurality of text passages, each of the text passages being associated with only one of the plurality of future events associated with the first document.

20. A computer program product for enabling searching for future events associated with a set of documents stored in a database on at least one memory device, the computer program product comprising at least one computer-readable storage medium having computer program instructions stored therein which are operable to cause one or more computing devices to:

identify one or more future events associated with each of the documents, at least some of the documents having a plurality of the future events associated therewith;

generate temporal data and a confidence value for each future event, the temporal data for each future event comprising a time segment, the confidence value representing a likelihood of occurrence for the corresponding future event and being determined with reference to text associated with the corresponding future event by identifying one or both of a verb tense or verb mood for the associated text, each future event corresponding to a tuple of the corresponding time segment and confidence value;

identify a text scope for at least some of the future events, the text scope corresponding to a subset of the document with which the corresponding future event is associated;

evaluate the text associated with the future events to generate event metadata for each future event;

index the subsets of the documents corresponding to each future event with the corresponding event metadata, and the corresponding tuple of the corresponding time segment and confidence value in the database; and

generate search results corresponding to selected ones of the documents in the database in response to queries which include references to future times, and to rank the search results in accordance with the confidence values associated with the future events associated with the selected documents.