Process and system for arrangement of documents

• US 5,745,893 A
• Filed: 11/30/1995
• Issued: 04/28/1998
• Est. Priority Date: 11/30/1995
• Status: Expired due to Term

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Electronic Data Systems Corporation (Perspecta, Inc.)
• Inventors: Thompson, Gregory J., Hill, Joe R.
• Primary Examiner(s): Black, Thomas G.
• Assistant Examiner(s): Coby, Frantz
• Application Number:
  US08/565,806
• Time in Patent Office: 880 Days
• Field of Search: 395/605, 395/606, 364/300, 364/401, 707/5, 707/6, 707/7, 707/3
• US Class Current: 715/200
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: Y10S 707/917 Text
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99937 Sorting

Abstract

According to one embodiment of the present invention, a system (10) for arrangement of documents (13) is provided. The system includes a document repository (12) storing a plurality of documents (13). An arrangement engine (14) is coupled to the document repository (12). The arrangement engine (14) is operable to build a plurality of symmetric matrices based upon a plurality of identified arrangements of the plurality of documents (13) in the document repository (12). Each matrix is structured such that each row and each column is associated with a document (13). The arrangement engine (14) is also operable: to fill each cell of each symmetric matrix with a relevance value representing a relevance between a document (13) associated with the row of the cell and a document (13) associated with the column of the cell, to generate a fitness value for each matrix based upon a magnitude and a position of each relevance value by analyzing each symmetric matrix using a fitness function (16), to identify an optimal arrangement of the plurality of documents (13) based upon the fitness value of each matrix, and to provide the optimal arrangement of documents (18) as system output.

Claims

1. A system for arrangement of documents, comprising:

a document repository storing a plurality of documents; and

an arrangement engine coupled to the document repository, the arrangement engine operable to;

build a plurality of symmetric matrices based upon a plurality of identified arrangements of the plurality of documents in the document repository, each symmetric matrix structured such that each row and corresponding column are associated with a particular document;

fill each cell of each symmetric matrix with a relevance value representing a relevance between a document associated with the row of the cell and a document associated with the column of the cell;

generate a fitness value for each symmetric matrix based upon a magnitude and a position of each relevance value by analyzing each symmetric matrix using a fitness function;

identify an optimal arrangement of the plurality of documents based upon the fitness value of each symmetric matrix; and

provide an optimal arrangement of documents as system output.

2. The system of claim 1, wherein the system is implemented on a computer system having a storage device, a memory and a processor.

3. The system of claim 1, wherein each relevance value is based upon a relevance measure between two documents based upon a log-likelihood ratio contrasting the hypothesis that the two documents are relevant to each other versus the hypothesis that the two documents are not relevant to each other.

4. The system of claim 3, wherein the fitness function used to generate a fitness value comprises a weighted fitness function.

5. The system of claim 1, wherein each document comprises text information.

6. A process for arrangement of documents, comprising:

building a plurality of symmetric matrices based upon a plurality of identified arrangements of a plurality of documents in a document repository, each symmetric matrix structured such that each row and corresponding column are associated with a particular document;

filling each cell of each symmetric matrix with a relevance value representing a relevance between a document associated with the row of the cell and a document associated with the column of the cell;

generating a fitness value for each symmetric matrix based upon a magnitude and a position of each relevance value by analyzing each symmetric matrix using a fitness function;

identifying an optimal arrangement of the plurality of documents based upon the fitness value of each symmetric matrix; and

providing the optimal arrangement of documents as process output.

7. The process of claim 6, wherein the process is implemented on a computer system having a magnetic storage device, a memory and a processor.

8. The process of claim 6, wherein filling each cell comprises filling each cell with a relevance value based upon a relevance measure between the two documents based upon a log-likelihood ratio contrasting the hypothesis that two documents are relevant to each other versus the hypothesis that the two documents are not relevant to each other.

9. The process of claim 6, wherein generating a fitness value comprises generating a fitness value using a fitness function comprising a weighted fitness function.

10. The process of claim 6, wherein building comprises building with respect to documents comprising text information.

11. A process for arrangement of documents and identification of sub-classes of documents, comprising:

building a plurality of symmetric matrices based upon a plurality of identified arrangements of a plurality of documents in a document repository, each symmetric matrix structured such that each row and corresponding column are associated with a particular document;

filling each cell of each symmetric matrix with a relevance value representing a relevance between a document associated with the row of the cell and a document associated with the column of the cell;

generating a fitness value for each symmetric matrix based upon a magnitude and a position of each relevance value by analyzing each symmetric matrix using a fitness function;

identifying an optimal arrangement of the plurality of documents based upon the fitness value of each symmetric matrix; and

identifying a plurality of sub-classes of documents from the optimal arrangement of documents.

12. The process of claim 11, wherein the process is implemented on a computer system having a magnetic storage device, a memory and a processor.

13. The process of claim 11, wherein filling each cell comprises filling each cell with a relevance value based upon a relevance measure between two documents based upon a log-likelihood ratio contrasting the hypothesis that the two documents are relevant to each other versus the hypothesis that the two documents are not relevant to each other.

14. The process of claim 11, wherein building comprises building with respect to documents comprising text information.

15. The process of claim 11, wherein generating a fitness value comprises generating a fitness value using a fitness function comprising a weighted fitness function.