Method and apparatus for maintaining and navigating a non-hierarchical personal spatial file system

• US 20050187931A1
• Filed: 04/25/2005
• Published: 08/25/2005
• Est. Priority Date: 11/06/2000
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Lenchner, Jonathan, Cofino, Thomas A.
• Granted Patent: US 8,515,959 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/22 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/3344 using natural language analysis

Abstract

A self-organizing personal file system is disclosed that evaluates the “importance” of terms and phrases in a document in a personal corpus relative to usage in a reference corpus. A personalized term weighting scheme assigns a weight to terms or phrases based on the frequency of occurrence of the corresponding term or phrase in a reference corpus. The personalized term weighting for a given term or phrase can be used to store and access documents containing the corresponding term or phrase in the spatial file system and provides coordinates in a spatial file system, for one or more documents containing the corresponding term or phrase. The location of a given document in a file space may be specified by the relative frequency distribution of the stems of its significant terms or phrases compared to the occurrence of such terms or phrases in a reference corpus. The employed term weighting scheme can be based on the fractional frequency of a term in a user'"'"'s document relative to the fractional frequency of the term in the reference corpus, or alternative term weighting schemes, such as entropy weightings and traditional tf×idf weightings can be employed. A spatial rendering system utilizes the personalized term weights as coordinates for each document in a spatial file system. The spatial rendering system positions landmarks in the visual file space indicating terms that are correlated to terms associated with axes of the file system.

Claims

1. A method for storing a document in a file system, comprising the steps of:

determining a term weight for terms appearing in said document, wherein said term weight is based on a frequency of occurrence of said terms in a reference corpus; and

storing said document with an indication of said term weights.

2. The method of claim 1, wherein said term weight is obtained by dividing a fractional frequency of said term in said document by a fractional frequency of said term in said reference corpus.

3. The method of claim 2, wherein said fractional frequency of said term in said document is the number of occurrences of the term in the document divided by the total number of terms in the document.

4. The method of claim 2, wherein said fractional frequency of said term in said reference corpus is the number of occurrences of the term in the reference corpus divided by the total number of words in the reference corpus.

5. The method of claim 2, further comprising the step of computing an overall document weight for said document.

6. The method of claim 5, wherein said overall document weight indicates the total information content of a document.

7. The method of claim 5, wherein said overall document weight is the sum of the term weights for all terms in the document.

8. The method of claim 1, wherein said term weights are stored in a relational database.

9. The method of claim 1, wherein said term weights are stored in an inverted index.

10. The method of claim 1, wherein said term weights are stored in an array.

11. The method of claim 1, wherein said term weights are stored in a flat file.

12. The method of claim 1, wherein said document can be coordinatized by said term weights.

13. The method of claim 1, wherein said document can be indexed by said term weights.

14. A method for identifying one or more documents in a file system that are related to one or more specified words, comprising the steps of:

storing each of said documents with an indication of a term weight for terms appearing in said corresponding document, wherein said term weight is based on a frequency of occurrence of said terms in a reference corpus; and

evaluating a distance between each of said documents and said one or more specified words, wherein said distance is based on said term weights.

15. The method of claim 14, wherein said distance is the square root of the sum of the squares of the differences between said one or more specified words and terms in said document, using said term weights.

16. The method of claim 14, wherein said term weight is obtained by dividing a fractional frequency of said term in said document by a fractional frequency of said term in said reference corpus.

17. The method of claim 16, wherein said fractional frequency of said term in said document is the number of occurrences of the term in the document divided by the total number of terms in the document.

18. The method of claim 16, wherein said fractional frequency of said term in said reference corpus is the number of occurrences of the term in the reference corpus divided by the total number of terms in the reference corpus.

19. A system for storing a document in a file system, comprising:

a memory that stores computer-readable code; and

a processor operatively coupled to said memory, said processor configured to implement said computer-readable code, said computer-readable code configured to;

determine a term weight for terms appearing in said document, wherein said term weight is based on a frequency of occurrence of said terms in a reference corpus; and

store said document with an indication of said term weights.

20. A system for identifying one or more documents in a file system that are related to one or more specified words, comprising:

a memory that stores computer-readable code; and

a processor operatively coupled to said memory, said processor configured to implement said computer-readable code, said computer-readable code configured to;

store each of said documents with an indication of a term weight for terms appearing in said corresponding document, wherein said term weight is based on a frequency of occurrence of said terms in a reference corpus; and

evaluate a distance between each of said documents and said one or more specified words, wherein said distance is based on said term weights.

21. An article of manufacture for storing a document in a file system, comprising:

a computer readable medium having computer readable code means embodied thereon, said computer readable program code means comprising;

a step to determine a term weight for terms appearing in said document, wherein said term weight is based on a frequency of occurrence of said terms in a reference corpus; and

a step to store said document with an indication of said term weights.

22. An article of manufacture for identifying one or more documents in a file system that are related to one or more specified words, comprising:

a computer readable medium having computer readable code means embodied thereon, said computer readable program code means comprising;

a step to store each of said documents with an indication of a term weight for terms appearing in said corresponding document, wherein said term weight is based on a frequency of occurrence of said terms in a reference corpus; and

a step to evaluate a distance between each of said documents and said one or more specified words, wherein said distance is based on said term weights.