Office correspondence storage and retrieval system

• US 4,358,824 A
• Filed: 12/28/1979
• Issued: 11/09/1982
• Est. Priority Date: 12/28/1979
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Glickman, David, Rosenbaum, Walter S., Repass, James T., Russell, Janet G.
• Primary Examiner(s): Zache, Raulfe B.
• Application Number:
  US06/107,994
• Time in Patent Office: 1,047 Days
• Field of Search: 364/200 MS File, 364/900 MS FILE, 364/300
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3341 using boolean model
• CPC Class Codes: G06F 16/345 Summarisation for human users
• CPC Class Codes: G06F 40/279 Recognition of textual entities
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/35 Discourse or dialogue representation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

A system that intelligently abstracts and archives a document for storage and interprets a free form user retrieval query to recall the document from the storage file. The system includes a method for automatically selecting keywords from the document using a parts of a speech directory. A method is given for weighing the importance or centrality of each keyword with respect to the document of its origin. Using the same logic paths, a free form query that describes the document in the same manner that it would have to be described to a secretary to "find" it in a filing cabinet, the system automatically determines the key matching terms and finds the archived document(s) with the greatest affinity.

Claims

1. A method for abstracting and archiving a document in machine readable form comprising the steps of:

(a) storing a dictionary of language terms commonly used in document preparation;

(b) appending codes to the language terms in said dictionary of language terms to identify selected parts of speech;

(c) comparing the language terms in an input document with the stored dictionary of language terms;

(d) selecting language terms from said input document which do not compare to the stored dictionary of language terms;

(e) selecting language terms from said input document which compare with language terms in said stored dictionary of language terms identified as selected parts of speech;

(f) coding the selected language terms with the identity of the input document; and

(g) storing the selected language terms for later recall.

2. The method of claim 1 further including the steps of accumulating a count for the number of times each of the selected language terms occurs in the input document and accumulating a count of the number of pages in the input document.

3. The method of claim 1 or claim 2 further including the step of appending to each selected language term a code indicating the position of occurrence of the selected language term in the input document.

4. A method for retrieving a document from storage in response to input language terms descriptive of the content of the document comprising the steps of:

(a) comparing each of the input language terms to stored document abstract files of language terms, each document abstract language term having associated with it a code identifying its part of speech, a count indicating its frequency of occurrence in the document, a count of the number of pages in the document, and an indicator of the position of occurrence of the term in the document;

(b) accumulating a retrieval record for each document abstract file composed of the language terms that compare equal;

(c) calculating a document retrieval value for each retrieval record using the part of speech code, frequency count, number of pages in the document, and position indicator for each language term in the retrieval record;

(d) increasing the document retrieval value for each retrieval record that includes a month and/or year; and

(e) selecting the document corresponding to the highest calculated retrieval value for output.

5. The method of claim 4 further including the step of selecting all documents whose calculated retrieval value is equal to or greater than a predetermined percentage of the highest calculated retrieval value.

6. A system for abstracting a document in machine readable form comprising:

means for storing a dictionary of language terms commonly used in document preparation, said language terms including a code identifying certain ones of said language terms as selected parts of speech;

means for receiving an input document of language terms in machine readable form, said input document including an identification code;

a memory;

control means connected to said means for storing, said means for receiving and said memory, including,

means for comparing the language terms of said input document to said dictionary of language terms,

first selecting means responsive to said means for comparing for selecting the language terms from said input document that compare unequal,

second selecting means responsive to said means for comparing for selecting the language terms from said input document that compare equal and are coded as selected parts of speech;

first counting means responsive to said first and second selecting means for counting the frequency of occurrence of each selected language term in the input document;

second counting means responsive to said means for receiving for counting the number of pages in the document;

means responsive to said first and second selecting means for calculating the position of occurrence of the selected language terms in the input document; and

means responsive to said first and second selecting means, said first and second counting means, and said means for calculating for storing in said memory a record of each selected language term including the document identification code, the language term, the selected part of speech code, the frequency of occurrence count, the count of pages in the document, and the position of occurrence code.

7. The system of claim 6 wherein said control means further includes means for comparing each selected language term from the input document to selected language terms currently stored in said memory, and means responsive to an equal compare for adding to the record of the selected language term stored in said memory the identification code of the input document, the frequency of occurrence count, and position of occurrence code for the selected language term, thereby eliminating the need for duplicate storage of the selected language term.

8. A system for retrieving a document from storage in response to an input query of language terms descriptive of the content of the document comprising:

a memory having stored therein language term records including the language term, identification codes of documents containing the language term, a selected parts of speech code, a frequency of occurrence count for the language term, a count of pages in each document, and a position of occurrence code for each document identification code in each language term record;

means for comparing the language terms of the input query to language term records stored in said memory;

means for accumulating a retrieval record for each document identification code of each language term that compares equal;

means responsive to said means for accumulating for calculating a document retrieval value for each retrieval record using the selected part of speech code, frequency of occurrence count, count of pages and position of occurrence code; and

means responsive to said means for calculating for outputting from memory the document whose identification code corresponds to the identification code for the highest calculated retrieval value.

9. The system of claim 8 wherein said means for calculating further includes means for increasing the document retrieval value for each retrieval record that includes a month that compares equal to a term in the input query and further increasing the document retrieval value for each record that includes a year that compares equal to a term in the input query.

10. The system of claim 8 or claim 9 wherein said means for calculating includes means calculating a percentage of the highest calculated retrieval value for each document identification code and said means for outputting further includes means for outputting all documents whose retrieval value exceeds a predetermined percentage of the highest calculated retrieval value.

11. The system of claim 10 wherein means for outputting further includes means for selecting documents for display in the descending order of the number of query terms that matched langugage term records for the document.