Temporal Document Sorter and Method

• US 20090049037A1
• Filed: 08/14/2008
• Published: 02/19/2009
• Est. Priority Date: 08/14/2007
• Status: Active Grant

• Current Assignee: John Nicholas and Kristin Gross Trust U/A/D April 13 2010
• Original Assignee: John Nicholas and Kristin Gross Trust U/A/D April 13 2010
• Inventors: Gross, John Nicholas
• Granted Patent: US 10,762,080 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24 Querying
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/95 Retrieval from the web
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9536 Search customisation based on social or collaborative filtering
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 30/00 Commerce
• CPC Class Codes: G06Q 30/0243 Comparative campaigns
• CPC Class Codes: G06Q 30/0251 Targeted advertisements

Abstract

Electronic documents are classified and compared according to their temporal qualities. The content of a document relating to an event is analyzed to identify temporal components. These components can be compared with corresponding counterparts in other documents to identify a relative temporal order. The invention can be used in environments such as automated news aggregators, search engines, and other electronic systems which compile information having temporal qualities.

Claims

1. A method of automatically classifying temporal characteristics of electronic documents with a computing system comprising:

a) determining at least a first topic of a first document with the computing system;

b) comparing said first document with at least one reference document associated with said first topic having a known temporal value with the computing system;

c) generating a first temporal value for said first document with the computing system based on said step b);

wherein said first temporal value is derived from content in said first document other than but not excluding date/time data specified for said first document.

2. The method of claim 1, wherein said at least one reference document is derived from statistical analysis of one or more training documents associated with said first topic.

3. The method of claim 1, wherein said at least one reference document is derived from analyzing a second document containing content which overlaps with content in said first document.

4. The method of claim 3 wherein said first document contains a first content which is a superset of a second content found in said second document.

5. The method of claim 3 wherein said first document contains first content which is a subset of second content found in said second document.

6. The method of claim 3 wherein said first document contains a first content which is related semantically to a portion of a second content found in said second document.

7. The method of claim 1, further including a step:

collecting said first document and a set of related documents automatically from one more Internet-based content sources.

8. The method of claim 1 wherein said first document is compared against a plurality of reference documents before being assigned said first temporal value.

9. The method of claim 1 wherein an alert is generated if said first temporal value exceeds a threshold.

10. The method of claim 9 wherein said first temporal value is verified by a human and/or a computing machine prior to assigning a permanent temporal value to said first document.

11. The method of claim 1, wherein a set of documents related to said first topic is automatically collected and temporally ranked.

12. The method of claim 11, wherein documents containing duplicate content are discarded in said set of document at least during display of the same over the Internet.

13. The method of claim 1, wherein a first source value of said first document is determined, and said first temporal value is modified based on said first source value.

14. The method of claim 1 wherein said first document is analyzed to determine a first content snippet which is predicted to change semantically with time.

15. The method of claim 1 wherein said first document is compared to said at least one reference document to identify a first content snippet which is predicted to change semantically with time.

16. The method of claim 14, wherein semantic variants and temporal derivatives of said first content are computed and stored for future evaluations of subsequent documents.

17. The document of claim 16, wherein predicted versions of temporal variations of said first document are computed.

18. The method of claim 14, wherein a subsequent document is analyzed with respect to determine if it contains semantic variants and/or temporally related derivatives of said first content.

19. The method of claim 1 wherein said steps are implemented as one or more computer software routines embodied in a tangible media and adapted to cause one or more computing systems to perform the operations recited therein.

20. A method of automatically sorting electronic documents by their temporal characteristics with a computing system comprising:

a) identifying at least a first event having first content embodied in a first electronic document with the computing system;

b) analyzing said first electronic document with the computing system to identify first temporal components;

wherein said first temporal components relate to a state of said first event;

c) analyzing a second electronic document with second content related to said at least first event with the computing system;

d) generating an output with the computing system identifying which of said first and second electronic documents contains content describing a more current state of said first event.

21. The method of claim 20 wherein said temporal components consist of selected text words and/or phrases.

22. The method of claim 20 wherein first and second temporal interpretations are generated for said first document and second document respectively to determine which contains said content relating to a more current state of said event.

23. The method of claim 20 further including a step:

determining second temporal components in said second document relating to said first event, which second temporal components are compared to said first temporal components to determine which document contains content relating to a more current state of said event.

24. The method of claim 20 further including the steps:

automatically locating and compiling a set of documents relating to said at least first event; and

sorting said documents according to their temporal characteristics relating to said first event.

25. The method of claim 20, further including a step:

generating an alert in human detectable form in response to said output of step (g) to inform a human operator of potential new content relating to said event.

26. The method of claim 20, further including steps:

determining a first source of first electronic document and a second source of said second electronic document; and

attributing respective confidence rankings to each of said first source and said second source, which confidence rankings are used in determining which of said electronic documents contains content relating to a more current state of said first event.

27. The method of claim 20 wherein said current state relates to an absolute time associated with said event.

28. The method of claim 20 wherein said content corresponds to a first electronic incarnation of previously unpublished factual information relating to said event.

29. The method of claim 20 wherein said content corresponds to a first electronic incarnation of previously unpublished factual information relating to a current state of said event.

30. The method of claim 20, further including steps:

determining a site of said first event; and

at least in part prioritizing collection of electronic documents relating to said first event from one or more sources geographically associated with said site.

31. The method of claim 30, further including a step:

weighting electronic documents from said site related sources with a higher score relative to other sources reporting on said first event.

32. The method of claim 20 further including a step:

validating an existence of said more current state by reference to one or more online electronic searches.

33. The method of claim 20 further including a step:

validating an existence of said more current state by reference to rankings provided by one or more online human reviewers of said first and second electronic documents.

34. The method of claim 33, wherein said online human reviewers are drawn at least in part from a geographical region associated with said event.

35. The method of claim 33, wherein said online human reviewers are drawn primarily from a controllable sized geographical region of a associated with said event.

36. The method of claim 20 wherein said steps are implemented as one or more computer software routines embodied in a tangible media and adapted to cause one or more computing systems to perform the operations recited therein.

37. A method of automatically sorting electronic documents by their temporal characteristics with a computing system comprising:

a. generating a first approximate temporal interpretation of a first electronic document relative to a state of an event;

b. wherein said first temporal interpretation is based on temporal characteristics of a set of electronic training documents;

c. using said first approximate temporal interpretation to identify a comparison set of documents for said first electronic document;

d. generating a second temporal interpretation of said first electronic document based on said first temporal interpretation and correlating such document with one or more second electronic documents having a third temporal interpretation value and taken from said comparison set.

38. The method of claim 37 wherein said steps are implemented as one or more computer software routines embodied in a tangible media and adapted to cause one or more computing systems to perform the operations recited therein.

39. A method of automatically sorting electronic documents by their temporal characteristics with a computing system comprising:

a. providing a set of reference temporal interpretation values;

wherein each of said reference temporal interpretation values correspond to a state of an event;

b. correlating a first electronic document with the computing system to identify a corresponding first temporal interpretation value;

c. comparing said first electronic document with a second electronic document having a known temporal interpretation value to determine a second temporal interpretation value;

d. generating a final temporal related value for said first electronic document based on said first interpretation value and said second interpretation value.

40. The method of claim 39 wherein said first electronic document is part of a batch of documents which are sorted into temporal order with respect to a particular event.

41. The method of claim 39 wherein said steps are implemented as one or more computer software routines embodied in a tangible media and adapted to cause one or more computing systems to perform the operations recited therein.