CA1241122A

CA1241122A - Office correspondence storage and retrieval system

Info

Publication number: CA1241122A
Application number: CA000363345A
Authority: CA
Inventors: David Glickman; James T. Repass; Walter S. Rosenbaum; Janet G. Russell
Original assignee: International Business Machines Corp
Current assignee: International Business Machines Corp
Priority date: 1979-12-28
Filing date: 1980-10-27
Publication date: 1988-08-23
Also published as: US4358824A; ES8206059A1; JPS5828616B2; EP0032194A1; JPS5850071A; ES498162A0; JPS5844536A; JPS6330648B2

Abstract

OFFICE CORRESPONDENCE STORAGE AND
RETRIEVAL SYSTEM

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 partial 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 descried 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.

Description

OFFICE CORRESPONDENCE STORAGE AND
RETRIEVAL SYSTEM

Background of the Invention ield Qf the Invention This invention relates to information storage and retrieval and more particularly to methods of automatic-ally abstracting, storing and retrieving documents using free form inquiry.

Description of the Prior Art In implementing a document storage and retrieval system, the practicality and utility of such a facility is governed by the ease that respective documents are cataloged into the system and the efficiency with which a user's request can be associated with the related 15document catalog representation (description). State of the art document storage and retrieval is based on manually selecting keyworas to represent a document in the system's catalog or index and then effecting retrieval by recalling from memory appropriate keyword terms and 20either automatically or manually searching the index for an "appropriate" level of match against the prestored keywords. Procedures have been developed in the prior art for abstracting documents and retrieving them based on ~eyword matchin~. One of the procedures requires the 25requestor to supply in a fixed format eertain details about the subject document such as: author, addressee, date and keywords or phrases. For retrieval, a summary sorted listing is prepared under each o~ the above headings. The requestor must discern the appropriate 30document by examining the entries under the retrieval information headings. No latitude is allowed in the z~

search clues. The search may be done by manual perusal or using data processing global find commands.
A second procedure stores all non trivial words (i.e., ignores articles and pronouns, etc~) in a document Sas a totally inverted file. The document/line/word position of origin is maintained in the catalog. Search of the database for retrieval is effected by the user supplying keywords based on the user's memory. The catalog is automatically searched with the added facility lOthat the user carl specify relations that must exist between the keywords as they exist in the original text (i.e., keyword 1 is before keyword 2, etc.). An example of such a system is the IsM*Data Processing Division product Storage and Info~mation Retrieval System, 15commonly called STAIRS.
A third ~ethod for document storage and retrieval is simply storing the document in machine reaaable form and searching all documents using a "global find" logic for each user supplied keyword. Tn theory and in prac-20tice for small data bases, the "g~obal find" can bereplaced by the user reviewing the documents verbatim as they are displayed on a CRT type device.
However, in all the above procedures for document storage and retrieval, the major intelligent burden for 25abstraction and retrieval association matching is put on the user. Where the system aids in abstraction or matc~ing, it i.s done at the cost of voluminous cataloging procedures, massive data processing burden and a struc-tured format is required for the user to communicate for 30retrieval with the system.
* Registered Trade Mark Summary of the Invention It has been discovered that all non-trivial corre--spondence is made topic specific by a relatively small number of messa~e speciaLization terms. These are the 5words that transform the "boiler plate" of business correspondence into the message that the author wishes to convey. These terms consist mainly of numerics, proper names, acronyms, nouns and single purpose adjec-tives. Any meaningful description of a document for lOquery purposes must contain at least some of these terms which give the document its particular meaning. This invention includes a technique for reliably locating the message specialization terms in a document and forming an abstract of the document using these terms. The 15technique utillzes the data storage technology disclosed in U.S. Patent 3,995,254 issued November 30, 1976 to w.
S. Rosenbaum, and assigned to the applicant herein, to store a dictionary of words for spelling verification, howe~er, other dictionary storage methodologies could ~Oalso be used. The specialization terms in the dictionary memory additionally have ~ data bit appended to them to indicate their status as a noun or single purpose adjec-tive. Numerics, proper names, and acronyms are not stored in the dictionary memory. The text of the docu-25ment is compared with the contents of the dictionarymemoxy and those words that compare to nouns and single purpose adjectives in the dictionary and those words (proper names, numerics, acronymsJ not found in the dictionary memory are accumulated to form an abstract of 30the document. Each word in the abstract is then stored in a word index file. Records in the word index file include the word, the identification code of the docu-ment(s) in which the word occurs, the number of times :r ~

the word occurs in each respective document, an indicator as to whethex the word is a numeric, proper name/acronym, noun/single purpose adjective, and an indicator as to whether the word occurs in the header, trailer, body or 5copy list of the document. (A single purpose adjective is a word whose primary use is adjectival, for example heavy, round, old, new, the colors red, blue, etc.) The words in an input query for retrieval of a document are compared against the word index file. Since some words lOin the word index file may occur in severa]. documents, weighing factors are accorded each word based on the information stored with the word in the word index file.
A score ls accumulated for each document that contains any of the words in the retrieval query and those docu-15ments with highest scores are presented to the user forreview.

Brief Descrw~p~ion of the Drawin~s FIG. 1 is a block diagram of system components in the document storage and retrie~al system.
FIG. 2 is a flow chart of the operation in abstract-ing and storing a document.
FIG. 3 is a flow chart of the operation of the system in retrieving a document in response to a user query.

D cription of the Preferred Embodiment Referring to FIG. 1 there is shown a block diagram of a document storage and retrieval system which includes a processor or CPU 10 o the general purpose type which is capable o decoding and executing instructions. The 30processor 10 is in two-way communication over bus 13 with a memory 14 containlng instructions which contr~l its operation and define the present invention. The processor 10 is also in two-way communication over bus 7 with memory 8 which contains a partial speech dictionary 5where all nouns and single purpose adjectives are so noted. The memory 8 contains no numerics, acronyms or proper na~es. The processor 10 is also in two-way communication over bus 11 with main memory 12 ~hich is used for storing the documents and key word index files.
lOThe instruction memory 14 and dictionary memory 8 may be of the read only storage or random access storage type, while the main memory 12 is of the random access storage type.
For document abstracting and archiving an input 15register 16 receives the text words from a source (not shown) over bus 17. The source may be any of various input devices including keyboard, magnetic tape reader, magnetic cards/ disk/diskette files, etc. Test words are presented to processor 10 by register 16 over bus 15 20for processing in accordance with instructions stored in instruction memory 14. The results of the processing (abstraction) performed on the text contents of register 16 are transmitted to memory 12 over bus 11.
For document retrieval, input register 16 receives 25the query text statement from a source (not shown) over hus 17. The source may be any of various input devices such as a keyboard, script table, or specially constituted touchtone pad. The query statement text is presented to processor 10 by register 16 over bus 15 for processing 30in accordance with instructions stored in instruction memory 14. The processor 10 under control of instructions from instruction memory 14 communicates with the contents of dictionary memory 8 over bus 7 and memory 12 over bus 11 to perform a document retrieval affinity evaluation AT9-79~010 ''~ 6 ~ 2~
on the contents of memory 12. The selected document(s) are transmitted from memory 12 over hus 11 and bus 9 to output reglster 18 and from output register 18 over bus 19 to a utilization device which may take various forms, Slncluding a display, printer or voicecoder, etc. The selected document(s) are then presented to the user for review.
The preferred embodiment of the present invention comprises a set of instructions or programs for control-lOling the document abstracting, archiving and querystatement affinity match for retrieval for the document storage and retrieval system shown in FIG. 1. Referring to FIG. 2 there is shown a flow chart of the programs for abstracting and archiving documents.
It i5 standard practice in data processing systems having on-line storage to assign each record stored a unique identifier code or number. This code is usually eight characters in length and does not contain informa-tion that is descriptive of the contents of the record 20because of the limited length. The identifier code is useful for accessing the records where the user is able to associate the identifier code with a particular record. However, this technique for locating a record become impractical where the data base is large and 25several users have access to the same records. A
record usually retains the same identifier code through-out its existence and modifications to the record replace the record in storage under the same identifier code.
The program for abstracting and archiving documents 30makes use of the identifier code by including it as part o~ the abstract record. When a docu,ment is entered into ~` the System, FIG. 2, the document identifier code or number for the document is read at block 20 and the word index files already stored in the system are compared to 7 ~ o~
determine if a match is found indicating that an abstract is currently stored for the document.

Document Abstraction Routine BEGINPROCEDURE(OCRS_ABSTRACT);
ENTER ABSTRACT, SAVE DOCUMENT NUMBER PARAMETER;
READ DOCUMENT ABSTRACT FILE RECORD FOR DOCUMENT NUMBER;
IF
RECORD FOUND
lOTHEN
CALL (DELETE ABSTRACT);
ENDIF;
WHILE
NOT END OF DOCVMENT

WHILE
NOT END OF PAGE
DO
GET NEXT LINE OF TEXT FROM THE DOCUMENT;
WHILE
MORE CHARACTERS EXIST ON THE LINE
; DO
GET NEXT WORD FROM THE LINE ~2 OR MORE
CONSECUTIVE CHARACTERS A-Z, 0-9, OR
');
IF
THE WORD IS "CC"
THEN
SET CC LINE NUMBER TO THE DOCUMENT
LINE NUMBER MINUS l;
ENDIF;
CALL tABSTRACT_PROCESS_WORD);

8 ~ l2.
ENDWHILE;
INCREMENT PAGE NUMBER BY l;
ENDWHILE;
INCREMENT DOCUMENT LINE NUMBER BY l;
SENDWHILE;
SET LAST BODY LINE COUNT TO THE LESSOR OF:
THE CC LINE NUMBER AND THE DOCUMENT LINE NUMBER;
DECREMENT THE LAST BODY LINE COUNT BY 4;
CALL (ABSTRACT_END_PROCESSING);
ENDPROCEDURE(OCRS_ABSTRACT):

Table 1 is the program routine for abstracting the document. If the document number (identifier code) is found to exist in the abstract file, the program routine branches to the deLete abstract routine of Table 2 which 15is shown as block 22 of the flow chart of FIG. 2.

Delete Abstract Subroutine BEGINPROCEDURE(DELETE_ABSTRACT);
ENTER DELETE ABSTRACT;

: NOT END OF DOCUMENT ABSTRACT RECORD
DO
GET THE NEXT ENTRY IN THE DOCUMENT ABSTRACT RECORD;
READ THE WORD INDEX RECORD FOR THE WORD;
WHILE
NOTE END OF WORD INDEX RECORD
DO
GET THE NEXT ENTRY IN THE WORD INDEX RECORD;
IF

AT9-79-0l0

2~, THE DOCUMENT NUMBER IN THE ENTRY IS THE
SAME AS THE DOCUMENT NUMBER FROM
THE DOCUMENT ABSTRACT RECORD
THEN
REMOVE THE ENTRY FROM THE WORD INDEX
RECORD;
IF
THERE ARE NOW NO ENTRIES IN THE WORD
INDEX RECORD
THEN
DELETE THE WORD INDEX RECORD FROM
THE FILE;
ELSE
REWRI~E THE WORD INDEX RECORD T~ THE
FILE;
E~DIF;
ENDIF:
; ENDWHILE;
ENDWHILE;
20DELETE THE DOCUMENT ABSTRACT RECORD FROM THE FILE;
ENDPROCEDURE(DELETE ABSTRACT);

The delete abstract subroutine of Table 2 deletes the abstract from memory by deleting occurrences of the words in the abstract from the word index file. The 25makeup of the word index file will be fully explained below.
~ ollowing deletion of the existing abstract from memory, or, if no words having the document number are stored in the word index file, the document is processed 30at block 23 to create an abstract. Referring to the program routine in Table 1, the next word in the document is ~ested to determine if the Carbon Copy (CC) list follows. If not, the program branches to abstract process word routine in Table 3 to determine lf the word should be included in the abstract for the document.

Abstract Process Word Subroutine BEGINPROCEDURE (ABSTRACT _PROCES S_WORD );
ENTER PROCESS WORD
INCREMENT DOCUMENT WORD COUNT BY 1;
LOOK THE WORD UP IN THE DICTIONARY;
IF
T~E WORD WAS FOUND IN THE DICTIONARY BUT NOT ~I,AGGED
AS A NOUN OR A SINGLE PURPOSE ADJECTIVE
THEN
IGNORE THIS WORD;
ELSE

THE WORD WAS FOUND IN THE DICTIONARY BUT
FLAGGED AS A NOUN OR A SINGLE PURPOSE
ADJECTIVE
THEN
FLAG THE WORD AS NORMAL;
ELSE
FLAG THE WORD AS ACRONYM;
ENDIF;
IF
THIS WORD HAS NOT BEEN FOUND PREVIOUSLY IN
THIS DOCUME~T
THEN
SAVE THIS WORD;
SAVE THE DOCUMENT LINE COUNT;
SET FREQUENCY COUNT FOR THIS WORD TO 1;
ELSE
INCREMENT FREQUENCY COUNT FOR THIS WORD BY ~;

?l ENDIF;
ENDIF;
EN~PROCEDURE(ABSTRACT_PROCESS_WORD);

~s was previously stated, the criteria for determining 5whether a word is included in the abstract is whether the word is determined to be a "message specialization term, i.e., a noun, single purpose adjective, proper name, acronym, or numeric. The program routine of Table

3 compares the word to the contents of dictionary memory 103 (FIG. 1). If the word is found in the dictionary memory but it is not a noun or single purpose adjective then the word is ignored. The decision as to whether a word in the dictionary is a noun or single purpose adjective is made at the time of preparation of the 15dictionary memory 8 and those words designated as nouns or single purpose adjectives have appended to them a code bit. If the word is determined to be a noun or single purpose adjective, a code bit or "flag" is added to the word to indicate as "normal". If the word is not 20in the dictionary then a code bit or "flag" is added to the word to indicate its status as acronym or proper name. Acronyms and proper names are considered to have more influence as message specialization terms than nouns and single purpose adjectives and thereforP are 25more useful for document retrieval as will be shown below. The Process Word routine of Table 3 controls the processor 10 to save only one copy of each abstract term ~or storage in the word index fiLe. However, the Process Word routine appends to the word the number of each line 30in the document where the word appears and a count of the number of times the word appears in the document.
As will be seen below for document retrieval, the fre-~uency of occurrence of the word in the document and the place of occurrence help determin~ the value of the word as a query term for retrieving the document.
Following completion of the Word Process subroutine control returns to the Abstract routine in Table 1 which 5repeats ~he routines for each word in the document. The Abstract routine accumulates a count for the number of pages in the document. Upon reaching the end of the document a count is calculated to determine the fifth line from the end of the body of the document and the 10Abstract End Processing subroutine of Table 4 is selected.

Abstract End Processing Subroutine BEGINPROCEDURE(ABSTRACT_END_PROCESSING);
ENTER END PROCESSING;
15CREATE A DOCUMENT ABSTRACT RECORD CONSISTING OF;
THE DOCUMENT NUMBER, THE DOCUMENT WORD COUNT, A~D
EACH WORD IN THE ABSTRACT;
WRITE THE DOCUMENT ABSTRACT RECORD TO THE FILE;
WHILE
MORE WORDS ARE LEFT TO PROCESS;
DO
READ THE WORD INDEX RECORD FOR THE WORD;
IF
THE RECORD WAS NOT FOUND

CREATE A WO~D INDEX RECORD CONSISTING OF:
THE WORD, THE NORMAL/ACRONYM/PROP~R NAME
FLAG, THE DOCUM~NT NUMBER, THE FREQUENCY
COUNT, AND A FLAG INDICATING IN HEADER/
TRAILER/CC LIST/BODY;
WRITE THE WORD INDEX RECORD TO THE FILE;

ELSE
ADD l'~IE DOCUMENT NUMBER, THE FREQUENCY COUN~, AND A FLAG INDICATING IN HEADER/TRAILER/CC
LIST/BODY TO THE RECORD;
REWRITE THE WORD INDEX RECORD TO THE FILE;
ENDIF;
ENDWHILE;
ENDPROCEDURE(ABSTRACT_END_PROCESSING);

The Abstract End Processing subroutine controls the 10processor 10 to create an abstract record which includes all words saved by the Process Word subroutine of ~able 3, a count of the number of words in the document and the document identifier code number. The Abstract End Processing subroutine also creates a Word Index Record 15for each word in the abstract record which includes the word, the "normal" or "acronym/proper name" code, the document number, the number of pages in the document, the frequency of occurrence of the word in the document, and a code indicating whether the word occurs in the 20header (first 10 lines), trailer (last 5 lines) or the copy list or body of the document. The words in the Word Index File are searched to determine if a record for the word already appears in the Word Index File. If it does then the record is updated by adding the document 25number, frequency count, and codes such that no duplicates of the word appear in the Word Index File. Following completion of the Abstract End Processing subroutine o Table 4 control returns to the Abstract routine oE ~able 1 which terminates the abstracting procedure.
3a To retrieve a document stored in the system, the requestor must enter a query for the document into the system. This may be done through a keyboard, for example.
The queries used with the preferred embodiment of this 5yStem can be a natural language statement or string of words that describes the item. The search argument is created by testing the query words against the word index file. In many cases, the words in the search 5argument will occur in the key word records (abstracts) of several documents. In order to provide better discrim-ination between contending documents, different weights are applied to different key words. Weighting criteria are applied according to these general rules:

101 - Matches on numeric key words are given greater weight than matches on alpha key words.
2 - Matches with key words that are proper names or acronyms are given greater weight than matches with nouns or single purpose adjectives that are found in the dictionary memory.
3 The weight assigned to a key word match is propor-tional to the number of times that the word occurred in the document divided by the log of the number of pages in the document.
204 - Matches with key words that occur in the first ten lines of the document are given greater weight than those of key words in the center of the body of text.
5 - Matches that occur with key words in the last five lines of text ~before any copy lists) are given more weight than matches with words in the center of the text, ~ut less weight than matches with words in the first te~ lines.
6 - The weight of a key word match is increased when that word is the name of a month or year.
307 - The weight Oe a key word match is inversely propor-tional to the number of documents in the entire file that contain that key word in the body of the document (excluding occurrences as part of the copy list).

15 ~qf~
The rationale behind these general rules is to give the greatest weight to those matches that involve key words that have the most narrowly specific meaning. It is assumed that specific names, numbers and dates have 5very specific meaning so they are weighed hea~ily. It is also assumed that the most specific items will be mentioned at the beginning or end of the correspondence.
Hence, words occurring in these regions are also given greater weight. An example of an expression that satis-lOfies the general rules is the following:

Match Value =
F +lO i+lo i+lo i 5Ei Hii,j log2 Di (1 25) i(l 25) i 15where: Fi j = number of times ith key word appears in jth document divided log2 of the number of pages in document.
Ai = binary indicator if ith key word is an acronym or proper name.
Ki ~ ~inary indicator if ith key word occurs in first 10 lines.
Li = binary indicator if ith key word is a numericO
Ei = binary indicator if ith key word occurs in last S lines.
Hi = binary indicator if ith key word occurs in the dictionary as a noun or single purpose adjective.
Mi = binary indicator if ith key word is a month.
Yi = binary indicator if ith key word is a year.
Di = number o~ documents that contai~ ith key word.

Referring to FIG. 3, a flow chart of the processing of a query for a document is shown. At block 30 the user query is input to the processor 10 (FIG. l) from input regi~ter 16 over bus 15. Tables 5, 6, and 7 show 5program routines for processing the user query according to the general rules stated above.

Query Routine BEGINPROCEDURE(OCRS_QUERY);
10ENTER QUERY;
WHILE
MORE QUERY LINES OF TEXT EXIST
DO
GET THE NEXT LINE OF QUERY TEXT:
WHILE
MORE CHARACTERS EXIST ON THE LINE
- DO
GE~ THE NEXT WO~D FROM THE LINE (2 OR MORE
CHARACTERS A-Z, 0-9, OR ');
READ THE WORD INDEX RECORD FOR THE QUERY WORD
IF
; WORD FOUND
THEN
CALL (QUERY_PROCESS_WORD);
ENDIF
E~DWHILE;
E~D~HILE;
CALL ~QUERY END PROCESSING);
ENDPROCEDURE(OCRS QUERY);

30The Query routine of Table S compares the ~uery words to the contents of the word index file as shown in block 31 of the flow diagram of FIG. 3. The query words that match the word index file are processed at block 32 of the f low diagram by the Query Word Process subroutine of Table 6.

Query Process Word Subroutine Detailed Logic BEGINPROCEDURE(QUERY_PROCESS_WORD);
ENTER PROCESS WORD;
IF

THEN
SET INDICATOR FOR YEAR IN QUERY;
ENDIF;
IF
THE WORD IS R MONTH
THEN
SET INDICATOR FOR MONTH IN QUERY;
ENDIF;
; IF
THE WORD IS NUMERIC
THEN
SET NUMBER WEIGHT TO 10;
ELSE
SET NUMBER WEIGHT TO 0;
25 ENDIF;
THEN
COUNT THE NUMBER OF DOCUMENTS CONTAINING THIS WORD;
COUNT THE NUMBBR OF DOCUMENTS WHERE THE WORD IS
NOT IN THE CC LIST;
IF
THE WORD INDEX RECORD IS FLAGGED AS AN ACRONYM/
PROPER NAME

18 ~ .~
THEN
SET ACRONYM/PROPER NAME WEIGHT TO 10;
ELSE
SET NORMAL WEIGHT TO 5;
S ENDIF;
WHILE
MORE DOCUMENT ENTRIES ARE IN THE WORD INDEX
RECORD
DO
GET THE NEXT DOCUMENT ENTRY FROM THE WORD
INDEX RECORD
IF
THE FLAG INDICATES THAT THE WORD OCCURRED
I2~ THE HEP.DER
THEN
SET HEADER WEIGHT TO 10;
ELSE
SET HEADER WEIGHT TO 0;
ENDIF;
IF
: THE FLAG INDICATES THAT THE WORD OCCURRED
IN THE TRAILER
THEN
SET TRAILER WEIGHT TO 5;
ELSE
SET TRAILER WEIGHT TO 0;
ENDIF;
IF
THE FLAG INDICATES THAT THE WORD OCCURRED
IN THE CC LIST
THEN
SET CC DIVIDE WEIGHT TO 99,999;
ELSE
SET CC DIVIDE WEIGHT TO l;

...... ,.. .. .. .. .. .. . _.

~L2~ 2~
"

ENDIF;
SET THE RETRIEVAL VALUE TO:
(ACRONYM/PROPER NAME WEIG~T -~ NUMBER
WEIGHT + NORMAL WEIGHT + HEADER WEIG~T +
S TRAILER WEIGHT + WORD FREQUENCY DIVIDED
BY THE LOG BASE 2 OF COUNT OF N~MBER OF
PAGES) DIVIDED BY THE LOG BASE 2 OF THE
COU~T OF DOCUMENTS NOT CONTAINING THE
WORD IN THE CC LIST;
DIVIDE THE RETRIEVAL VALUE BY THE CC DIVIDE
WEIGHT;
IF
THIS DOCUMENT HAS NOT BEEN ANALYZED YET
IN T~IS QUERY
THEN
SAVE THE DOCUMENT NUMBER;
SAVE THE RETRIEVAL VALUE;
ELS~
INCREMENT THE DOCUMENTS RETRIEVAL VALUE
BY THE NEW RETRIEVAL VALUE;
ENDIF;
ENDW~ILE;
ENDPROCEDURE(QUERY_P~OCESS_WORD);

Each ~uery word is tested ta determine if it is a month, 25year, numeric, acronym or normal (nou~ or single purpose adjective). The subroutine of Table 6 also adds weight-ing factors if the indicators in the word index file ; show the ward occurs in the first ten lines ~Header) af the document, last five lines (Trailer) of the document, 300r occurs more than once in the document. The value of the word is reduced if it occurs in the copy list of the document or occurs in more than one document, An ov~,rall calculation of value for each word is c'alcula-ted and a 20 ~ q~
total value for all query words that match wo~ds in the word index file for each document number having any matches is accumulated. The steps of calculating the retrieval value for words and the retrieval value for 5documents are shown in block 33 and 34 of FIG. 3.
Following processing of all words in the query, the Query routine of Table 5 branches to the Month/Year Evaluation subroutine of Table 7.

Query Month/Year Evaluation ~EGINPROCEDURE~QUERY_END_PROCESSING);
ENTER END PROCESSING;
IF
THERE WAS A YEAR ~ENTIONED IN THE QUERY

INCREMENT THE RETRrEVAL VALUE OF EACH DOCUMENT
THAT DID CONTAIN THE YEAR BY 20%;
ENDIF;
IF
THERE WAS A MONTH MENTIONED IN THE QUERY
THEN
INCREMENT THE RETRIEVAL VALUE OF EACH DOCUMENT
THAT DID CONTAIN THE MONTH BY 20%;
ENDIF;

RETRIEVAL VALUE IS WITHIN 25% OE THE ~IGHEST
RETRIEVAL VALUE;
SORT TEIIS LIST BY THE NUMBER OF WO~DS FROM THE QUERY
ACTUALLY OCCURRING IN THE DOCUMENT;
300UTPUT THE DOCUMENTS;
ENDPROCEDURE(QUERY_END PROCESSING);

o~

The subroutine of Table 7 increases the retrieval value for each document that contains a year and/or month that matches a y~ar and/or month in the query. The subroutine of Table 7 then controls the processor 10 to output 5those documents from main memory 12 to output register 18 whose retrieval value is within 25 percent of the highest retrieval value calculated. Control is then returned to the Query routine of Table 5 which terminates the query procedure.
While the invention has been shown and described with reference to a specific set of computer instructions and retrieval weighting values, it will be understood by those skilled in the art that the spirit of this inven-tion can be implemented in other computer languages and 15the set of document retrieval weighting factors can ~e modified without avoiding the scope of the invention claimed herein.

. . . _ _ .

Claims

The embodiments of the invention in which an exclusive property or privilege is claime are defined as follows:

1. A method for abstracting, archiving and retrieving a document in machine readable form comprising the steps of:
(a) establishing a dictionary memory of preselected first specialization terms;
(b) comparing the text of said document with the terms in said dictionary memory;
(c) combining first specialization terms corresponding to terms in said text with second specialization terms, not entered in said dictionary memory, to form an abstract of said document;
(d) storing each word of the said abstract in a word index file;
retrieval of said documents being accomplished by:
(e) comparing retrieval query words with said word index file and selecting those documents having abstracts containing said query words.

2. A method as defined in claim 1 wherein said first specialization terms comprise nouns and single-purpose adjectives.

3. A method as defined in claim 2 wherein said second specialization terms comprise numerics, acronyms and proper names.

4. A method as defined in claim 3 wherein weighing factors are accorded each word stored in the word index file to effectively reduce the number of documents selected in response to the entry of a retrieval query.