Document retrieval system

• US 5,020,019 A
• Filed: 05/25/1990
• Issued: 05/28/1991
• Est. Priority Date: 05/29/1989
• Status: Expired due to Fees

• Current Assignee: Ricoh Company Limited
• Original Assignee: Ricoh Company Limited
• Inventors: Ogawa, Yasushi
• Primary Examiner(s): Fleming, Michael R.
• Assistant Examiner(s): Ray, Gopal C.
• Application Number:
  US07/528,490
• Time in Patent Office: 368 Days
• Field of Search: 364/200 MS File, 364/900 MS File, 364/419, 364/513
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3341 using boolean model
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: Y10S 706/90 Fuzzy logic
• CPC Class Codes: Y10S 706/934 Information retrieval or Information management
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99937 Sorting

Abstract

A document retrieval system which includes a keyword connection table making section, a document accuracy calculating section, a document sorting section and a learning control section. The document accuracy calculating section calculates a document accuracy for each of the output documents in a prescribed manner by reference to a keyword connection table file. The document sorting section sorts the output documents in downward sequential order of the document accuracy. The learning control section serves to modify the weight of each keyword connection in a prescribed manner after the sorted output documents are given responsive to a query by a user, allowing the user make an evaluation on whether each document accuracy of the output documents is in conformity with the query. The document retrieval system is capable of providing the user with multiple choices from a numerical value between 0 and 1 in terms of a real number in making an evaluation on whether each document accuracy of the output documents is actually in conformity with the query.

Claims

1. A document retrieval system comprising:

keyword connection table making means for making a keyword connection table which describes a weight of each keyword connection among a first set of keywords from each of documents registered in a database file, and, upon request, the keyword connection table making means acting to update a weight of a keyword connection pre-recorded on a keyword connection table file to make a new keyword connection table containing a modified weight of the keyword connection;

document selection means including means for selecting output documents from the database file by reference to the keyword connection table file, the output documents being in conformity with a query containing a second set of keywords initially given by a user;

document accuracy calculating means for calculating a document accuracy which is obtained with a weight of each keyword connection between a third set of keywords from a document out of the output documents and the second set of keywords from the query when the user modifies a document accuracy of the document, the document accuracy calculating means computing a document accuracy for each of the output documents in a prescribed manner by reference to the keyword connection table;

document sorting means for sorting the output documents in downward sequential order of the document accuracy computed by the document accuracy calculating means;

learning control means for modifying a weight of each keyword connection previously assigned between the third set of keywords from the document and the second set of keywords from the query in a prescribed manner when the user makes an evaluation on whether the document accuracy of the document previously computed by the document accuracy calculating means is in conformity with the query after the output documents are aligned by the document sorting means in downward sequential order of the document accuracy previously computed by the document accuracy calculating means; and

means for providing the user with multiple choices of a numerical value between 0 and 1 for each document accuracy of the output documents when the suer inputs an evaluation value to the document retrieval system after making the evaluation on whether the document accuracy of each of the output documents is in conformity with the query, the multiple choices each including binary 0 and 1 representing two distinctly opposed evaluations made by the user with respect to the query, or 0 being not in conformity with the query and 1 being in conformity with the query, the multiple choices each further including any intermediate numerical value between 0 and 1 representing an ambiguous evaluation, made by the user, that cannot be represented by the two distinctly opposed evaluation only.

2. The document retrieval system of claim 1, wherein the document selection means includes:

the document accuracy calculating means,

the document sorting means, and

the learning control means.

3. The document retrieval system of claim 2, wherein the document selection means further includes:

request handling means for handling a request given by the user from a keyboard, the request handling means acting to transfer the query containing the second set of the keywords to the document accuracy calculating means.

4. The document retrieval system as claimed in claim 3, wherein the document selection means further includes:

display control means for allowing retrieval results to be displayed on a display device, the document sorting means acting to transfer the output documents, aligned in downward sequential order of the document accuracy, to the display control means so that the user can look up each of the output documents accompanying a document accuracy on the display device.

5. The document retrieval system of claim 1, further comprising:

request handling means for handling a request from the user to transfer the query containing the second set of the keywords to the document accuracy calculating means; and

display control means for allowing retrieval results to be displayed on a display device;

wherein the document sorting means transfers the output documents, aligned in downward sequential order of the document accuracy, to the display control means so that the user can look up each of the output documents accompanying a document accuracy on the display device; and

wherein the keyword connection table making means, when a request is made by the user, updates the keyword connection table file to make small a difference between a document accuracy initially calculated by the document accuracy calculating means and an evaluation value subsequently given by the user.

6. The document retrieval system of claim 1, wherein:

the document accuracy calculating means computes a document accuracy B_i of a document (i) out of the output documents by calculating an algebraic sum of weights W_jk of keyword connections between a set of keywords KL(i) from the document out of the output documents and a set of keywords QUERY from the query given by the user, the algebraic sum of weights W_jk being represented by the formula;

##EQU8## where j is a keyword element among the set of keywords KL(i) and k is a keyword element among the set of keywords QUERY.

7. The document retrieval system of claim 1, wherein the document selection means includes:

request handling means for handling a request given by the user from a keyboard, the request handling means transferring the query containing the second set of the keywords to the document accuracy calculating means.

8. The document retrieval system of claim 1, wherein the document selection means includes:

display control means for allowing retrieval results to be displayed on a display device, the document sorting means transferring the output documents, aligned in downward sequential order of the document accuracy, to the display control means so that the user can look up each of the output documents accompanying a document accuracy on the display device.