Determining The Depths Of Words And Documents

• US 20090094209A1
• Filed: 10/01/2008
• Published: 04/09/2009
• Est. Priority Date: 10/05/2007
• Status: Abandoned Application

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Marvit, David L., Stergiou, Stergios, Jain, Jawahar
• Application Number:
  US12/243,012
• Publication Number:
  US 20090094209A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: G06F 40/242 Dictionaries

Abstract

In one embodiment, determining a document depth includes accessing a record that describes documents. The record records affinities associated with the documents. A document depth for a document is determined from the affinities. A document depth analysis may be performed using the document depth. In one embodiment, determining a word depth includes accessing a record that describes the affinities of words. A word depth is determined for a word from the affinities.

Claims

1. A method comprising:

accessing a record stored in one or more tangible media, the record describing a plurality of documents, a document comprising a plurality of words, the record recording a plurality of affinities associated with the plurality of documents;

determining a document depth for each document of at least a subset of the plurality of documents according to the plurality of affinities;

performing a document depth analysis according to the document depth to yield a result; and

reporting the result of the document depth analysis.

2. The method of claim 1:

the plurality of affinities further comprising a plurality of average affinities of the plurality of words.

3. The method of claim 1:

the record further comprising a document affinity graph comprising the plurality of affinities, the document affinity graph comprising a plurality of nodes, each node representing a document; and

the determining the document depth for the each document further comprising;

applying a link analysis to the document affinity graph to determine a popularity of at least a subset of the plurality of nodes;

determining that the each document is deeper if the each document is represented by a less popular node; and

determining that the each document is shallower if the each document is represented by a more popular node.

4. The method of claim 1, the performing the document depth analysis according to the document depth to yield the result further comprising:

receiving a search query, the search query comprising a document depth request; and

retrieving a set of documents of the plurality of documents that satisfy the search query and the document depth request.

5. The method of claim 1, the performing the document depth analysis according to the document depth to yield the result further comprising:

facilitating a graphical display of the document depth for the each document, the graphical display comprising one or more elements selected from a set consisting of a graphical indicator, a graphical modification, a depth slider, and a depth graph.

6. The method of claim 1, the performing the document depth analysis according to the document depth to yield the result further comprising:

receiving a search query;

retrieving a set of documents of the plurality of documents that satisfy the search query; and

sorting the set of documents.

7. The method of claim 1:

the plurality of documents associated with one or more users; and

the determining the document depth for the each document further comprising;

determining a user depth of the one or more users from the plurality of documents.

8. The method of claim 1:

the plurality of documents associated with one or more users;

the determining the document depth for the each document further comprising;

determining a user depth of the one or more users according to a theme from the plurality of documents.

9. One or more computer-readable tangible media encoding software operable when executed to:

access a record stored in one or more tangible media, the record describing a plurality of documents, a document comprising a plurality of words, the record recording a plurality of affinities associated with the plurality of documents;

determine a document depth for each document of at least a subset of the plurality of documents according to the plurality of affinities;

perform a document depth analysis according to the document depth to yield a result; and

report the result of the document depth analysis.

10. The computer-readable tangible media of claim 9:

the plurality of affinities further comprising a plurality of average affinities of the plurality of words.

11. The computer-readable tangible media of claim 9:

the record further comprising a document affinity graph comprising the plurality of affinities, the document affinity graph comprising a plurality of nodes, each node representing a document; and

the computer-readable tangible media further operable to determine the document depth for the each document by;

applying a link analysis to the document affinity graph to determine a popularity of at least a subset of the plurality of nodes;

determining that the each document is deeper if the each document is represented by a less popular node; and

determining that the each document is shallower if the each document is represented by a more popular node.

12. The computer-readable tangible media of claim 9, further operable to perform the document depth analysis according to the document depth to yield the result by:

receiving a search query, the search query comprising a document depth request; and

retrieving a set of documents of the plurality of documents that satisfy the search query and the document depth request.

13. The computer-readable tangible media of claim 9, further operable to perform the document depth analysis according to the document depth to yield the result by:

facilitating a graphical display of the document depth for the each document, the graphical display comprising one or more elements selected from a set consisting of a graphical indicator, a graphical modification, a depth slider, and a depth graph.

14. The computer-readable tangible media of claim 9, further operable to perform the document depth analysis according to the document depth to yield the result by:

receiving a search query;

retrieving a set of documents of the plurality of documents that satisfy the search query; and

sorting the set of documents.

15. The computer-readable tangible media of claim 9:

the plurality of documents associated with one or more users; and

the computer-readable tangible media further operable to determine the document depth for the each document by;

determining a user depth of the one or more users from the plurality of documents.

16. The computer-readable tangible media of claim 9:

the plurality of documents associated with one or more users;

the computer-readable tangible media further operable to determine the document depth for the each document by;

determining a user depth of the one or more users according to a theme from the plurality of documents.

17. A method comprising:

accessing a record stored in one or more tangible media, the record describing a plurality of words, the record comprising a plurality of affinities of the plurality of words;

determining a word depth for each word of the plurality of words according to the plurality of affinities to yield a plurality of word depths; and

reporting the plurality of word depths.

18. The method of claim 17:

the plurality of affinities further comprising a plurality of average affinities, an average affinity indicating a depth of an associated word; and

the determining the word depth for the each word of the plurality of words further comprising;

determining that the each word is deeper if the each word has a lower average affinity; and

determining that the each word is shallower if the each word has a higher average affinity.

19. The method of claim 17:

the record comprising a plurality of clusters generated from the plurality of affinities, the plurality of clusters comprising the plurality of words; and

the determining the word depth for the each word of the plurality of words further comprising;

determining that the each word is deeper if the each word belongs to fewer, smaller clusters; and

determining that the each word is shallower if the each word belongs to more, larger clusters.

20. The method of claim 17:

the record comprising an affinity graph comprising the plurality of affinities, the affinity graph comprising a plurality of nodes, each node representing a word; and

the determining the word depth for the each word of the plurality of words further comprising;

applying a link analysis to the affinity graph to determine a popularity of each node of the affinity graph;

determining that the each word is deeper if the each word is represented by a less popular node; and

determining that the each word is shallower if the each word is represented by a more popular node.

21. One or more computer-readable tangible media encoding software operable when executed to:

access a record stored in one or more tangible media, the record describing a plurality of words, the record comprising a plurality of affinities of the plurality of words;

determine a word depth for each word of the plurality of words according to the plurality of affinities to yield a plurality of word depths; and

report the plurality of word depths.

22. The computer-readable tangible media of claim 21:

the plurality of affinities further comprising a plurality of average affinities, an average affinity indicating a depth of an associated word; and

the computer-readable tangible media further operable to determine the word depth for the each word of the plurality of words by;

determining that the each word is deeper if the each word has a lower average affinity; and

determining that the each word is shallower if the each word has a higher average affinity.

23. The computer-readable tangible media of claim 21:

the record comprising a plurality of clusters generated from the plurality of affinities, the plurality of clusters comprising the plurality of words; and

the computer-readable tangible media further operable to determine the word depth for the each word of the plurality of words by;

determining that the each word is deeper if the each word belongs to fewer, smaller clusters; and

determining that the each word is shallower if the each word belongs to more, larger clusters.

24. The computer-readable tangible media of claim 21:

the record comprising an affinity graph comprising the plurality of affinities, the affinity graph comprising a plurality of nodes, each node representing a word; and

the computer-readable tangible media further operable to determine the word depth for the each word of the plurality of words by;

applying a link analysis to the affinity graph to determine a popularity of each node of the affinity graph;

determining that the each word is deeper if the each word is represented by a less popular node; and

determining that the each word is shallower if the each word is represented by a more popular node.

25. A method comprising:

accessing a record stored in one or more tangible media, the record describing a document comprising a plurality of words;

determining an average word depth of a selected set of two or more words of the plurality of words of the document; and

calculating a document depth of the document from the average word depth.

26. The method of claim 25, the selected set comprising two or more essential words of the document.

27. The method of claim 25, the selected set comprising the deepest X percent words, where X is 50 or greater.

28. The method of claim 25, the selected set comprising the deepest X percent words, where X is 50 or less.

29. The method of claim 25, the selected set excluding P percent of a plurality of standard grammar words of the document, where P is 50 or greater.

30. The method of claim 25, the selected set excluding P percent of a plurality of standard grammar words of the document, where P is 50 or less.

31. The method of claim 25, the selected set excluding Q percent of a plurality of stop words of the document, where Q is 50 or greater.

32. The method of claim 25, the selected set excluding Q percent of a plurality of stop words of the document, where Q is 50 or less.

33. One or more computer-readable tangible media encoding software operable when executed to:

access a record stored in one or more tangible media, the record describing a document comprising a plurality of words;

determine an average word depth of a selected set of two or more words of the plurality of words of the document; and

calculate a document depth of the document from the average word depth.

34. The computer-readable tangible media of claim 33, the selected set comprising two or more essential words of the document.

35. The computer-readable tangible media of claim 33, the selected set comprising the deepest X percent words, where X is 50 or greater.

36. The computer-readable tangible media of claim 33, the selected set comprising the deepest X percent words, where X is 50 or less.

37. The computer-readable tangible media of claim 33, the selected set excluding P percent of a plurality of standard grammar words of the document, where P is 50 or greater.

38. The computer-readable tangible media of claim 33, the selected set excluding P percent of a plurality of standard grammar words of the document, where P is 50 or less.

39. The computer-readable tangible media of claim 33, the selected set excluding Q percent of a plurality of stop words of the document, where Q is 50 or greater.

40. The computer-readable tangible media of claim 33, the selected set excluding Q percent of a plurality of stop words of the document, where Q is 50 or less.