Document retrieval system recognizing types and values of numeric search conditions

• US 7,509,314 B2
• Filed: 03/04/2005
• Issued: 03/24/2009
• Est. Priority Date: 03/05/2004
• Status: Expired due to Fees

• Current Assignee: OKI Electric Industry Company Limited
• Original Assignee: OKI Electric Industry Company Limited
• Inventors: Hamaguchi, Yoshitaka
• Primary Examiner(s): Vo, Tim T
• Assistant Examiner(s): OWYANG, MICHELLE N
• Application Number:
  US11/071,111
• Publication Number:
  US 20050198027A1
• Time in Patent Office: 1,481 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing
• CPC Class Codes: G06F 16/93 Document management systems
• 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

Abstract

A document retrieval system has an index information generator that generates index information about the frequency of occurrence of non-numeric terms in documents, and index information about the numeric values of numeric terms and the types of quantities they represent. Numeric terms received as search conditions are processed by a numeric information generator that generates search information indicating the type and numeric value of each numeric term. The index information generator and numeric information generator preferably convert numeric values to a standard notation. Numeric search conditions can then be used appropriately, flexibly, and conveniently to obtain useful retrieval results.

Claims

1. A document retrieval system for retrieving a document from a set of documents according to one or more terms received as search conditions submitted by a user, the terms that can be received including both numeric terms, which include numeric values, and non-numeric terms, which do not include numeric values, the document retrieval system comprising:

a first processor for generating at least one item of index information for each document in the set of documents, the index information generated for the documents including both numeric term index information and non-numeric term index information, the non-numeric term index information including information about frequency of occurrence of non-numeric terms in the document, the numeric term index information including information about types and numeric values of numeric terms in the document;

a memory unit for storing the index information generated by the first processor;

a second processor for examining each numeric term received as a search condition, determining its type, and generating numeric search information indicating the type and numeric value of the numeric term;

a third processor for using the non-numeric term index information generated by the first processor to calculate, for each document in the set of documents including a non-numeric term received as a search condition, a first numeric quantity having a single numeric value indicating a goodness of fit between the non-numeric term and the document;

a fourth processor for using the numeric term index information generated by the first processor and the numeric search information generated by the second processor from a numeric term received as a search condition to calculate, for each document in the set of documents including a numeric term of the same type as the numeric term received as the search condition, by comparing numeric values of the numeric terms, a second numeric quantity having a single numeric value indicating a goodness of fit between the numeric term received as the search condition and the document, closer numeric values indicating better fits;

a fifth processor for calculating, for a document in the set of documents, a third numeric quantity indicating a goodness of fit between the document and the one or more terms received as search conditions, the third numeric quantity being calculated from the first and second numeric quantities; and

an output unit for sending information for displaying a list of documents from the set of documents to the user according to the third numeric quantity.

2. The document retrieval system of claim 1, wherein the type of a numeric term indicates the type of quantity represented by the numeric value of the numeric term.

3. The document retrieval system of claim 1, further comprising wherein the index information is stored in a table in which frequencies of non-numeric terms and numeric values of numeric terms are both stored together in a single column.

4. The document retrieval system of claim 1, wherein the third numeric quantity is a sum of the first and second numeric quantities obtained for the document from the one or more terms received as search conditions.

5. The document retrieval system of claim 1, further comprising an output unit presenting documents for which the third numeric quantity exceeds a predetermined quantity.

6. The document retrieval system of claim 1, wherein the numeric term index information generated for a document in the set of documents includes, for each type of numeric term appearing in the document, all numeric values of numeric terms of said type appearing in the document.

7. The document retrieval system of claim 1, wherein the numeric term index information generated for a document in the set of documents includes, for each type of numeric term appearing in the document, a minimum numeric value appearing in the document and a maximum numeric value appearing in the document.

8. The document retrieval system of claim 7, wherein the numeric term index information generated for a document in the set of documents also includes, for each type of numeric term appearing in the document, a median value appearing in the document.

9. The document retrieval system of claim 1, wherein the numeric information generator and the index information generator convert numeric values to a standard notation.

10. The document retrieval system of claim 1, wherein each of the first processor, the second processor, the third processor, fourth processor, and the fifth processor separately comprises a memory and a computing device, the memory storing instructions executed by the computing device.

11. The document retrieval system of claim 1, further comprising a computing device, wherein each of the first processor, the second processor, the third processor, and the fourth processor and the fifth processor separately comprises a memory storing instructions executed by the computing device.

12. The document retrieval system of claim 1, wherein the fourth processor, in deriving the second numeric quantity, calculates a ratio between numeric values of the numeric term received as the search condition and the numeric term of the same type as the numeric term received as the search condition, appearing in the document.

13. A retrieval execution apparatus for use in a system for retrieving a document from a set of documents according to one or more terms received as search conditions submitted by a user, the terms that can be received including both numeric terms, which include numeric values, and non-numeric terms, which do not include numeric values, the retrieval execution apparatus having:

a first processor for generating at least one item of index information for each document in the set of documents, the index information generated for the documents including both numeric term index information and non-numeric term index information, the non-numeric term index information including information about frequency of occurrence of non-numeric terms in the document, the numeric term index information including information about types and numeric values of numeric terms in the document;

a second processor for examining each numeric term received as a search condition, determining its type, and generating numeric search information indicating the type and numeric value of the numeric term;

a third processor for using the non-numeric term index information generated by the index information generator to calculate, for each document in the set of documents including a non-numeric term received as a search condition, a first numeric quantity having a single numeric value indicating a goodness of fit between the non-numeric term and the document;

a fourth processor for using the numeric term index information generated by the first processor and the numeric search information generated by the second processor from a numeric term received as a search condition to calculate, for each document in the set of documents including a numeric term of the same type as the numeric term received as the search condition, by comparing numeric values of the numeric terms, a second numeric quantity having a single numeric value indicating a goodness of fit between the numeric term received as the search condition and the document, closer numeric values indicating better fits;

a fifth processor for calculating, for a document in the set of documents, a third numeric quantity indicating a goodness of fit between the document and the one or more terms received as search conditions, the third numeric quantity being calculated from the first and second numeric quantities; and

an output unit for sending information for displaying a list of documents from the set of documents to the user according to the third numeric quantity.

14. The retrieval execution apparatus of claim 13, wherein the type of a numeric term indicates the type of quantity represented by the numeric value of the numeric term.

15. A method of retrieving a document from a set of documents according to one or more terms received as search conditions submitted by a user, the terms that can be received including both numeric terms, which include numeric values, and non-numeric terms, which do not include numeric values, the method comprising:

generating at least one item of index information for each document in the set of documents, the index information generated for the documents including both numeric term index information and non-numeric term index information, the non-numeric term index information including information about frequency of occurrence of non-numeric terms in the document, the numeric term index information including information about types and numeric values of numeric terms appearing in the document; and

examining each numeric term received as a search condition, determining its type, and generating numeric search information indicating the type and numeric value of the numeric term;

using the non-numeric term index information to calculate, for each document in the set of documents including a non-numeric term received as a search condition, a first numeric quantity having a single numeric value indicating a goodness of fit between the non-numeric term and the document; and

using the numeric term index information and the numeric search information generated from a numeric term received as a search condition to calculate, for each document in the set of documents including a numeric term of the same type as the numeric term received as the search condition, by comparing numeric values of the numeric terms, a second numeric quantity having a single numeric value indicating a goodness of fit between the numeric term received as the search condition and the document, closer numeric values indicating better fits;

calculating, for a document in the set of documents, a third numeric quantity having a single numeric value indicating a goodness of fit between the document and the one or more terms received as search conditions, the third numeric quantity being calculated from the first and second numeric quantities; and

sending information for displaying a list of documents from the set of documents to the user according to the third numeric quantity.

16. A machine-readable medium storing a program for retrieving a document from a set of documents according to one or more terms received as search conditions submitted by a user, the search terms that can be received including both numeric terms, which include numeric values, and non-numeric terms, which do not include numeric values, the program being executable by computing device, the program including:

instructions for generating at least one item of index information for each document in the set of documents, the index information generated for the documents including both numeric term index information and non-numeric term index information, the non-numeric term index information including information about frequency of occurrence of non-numeric terms in the document, the numeric term index information including information about types and numeric values of numeric terms appearing in the document;

instructions for examining each numeric term received as a search condition, determining its type, and generating numeric search information indicating the type and numeric value of the numeric term;

instructions for using the non-numeric term index information to calculate, for each document in the set of documents including a non-numeric term received as a search condition, a first numeric quantity having a single numeric value indicating a goodness of fit between the non-numeric term and the document; and

instructions for using the numeric term index information and the numeric search information generated from a numeric term received as a search condition to calculate, for each document in the set of documents including a numeric term of the same type as the numeric term received as the search condition, by comparing numeric values of the numeric terms, a second numeric quantity having a single numeric value indicating a goodness of fit between the numeric term received as the search condition and the document, closer numeric values indicating better fits;

instructions for calculating, for a document in the set of documents, a third numeric quantity having a single numeric value indicating a goodness of fit between the document and the one or more terms received as search conditions, the third numeric quantity being calculated from the first and second numeric quantities; and

instructions for sending information for displaying a list of documents from the set of documents to the user according to the third numeric quantity.