Method and apparatus for classifying document information

• US 5,832,470 A
• Filed: 09/29/1995
• Issued: 11/03/1998
• Est. Priority Date: 09/30/1994
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Morita, Takako, Hashimoto, Tetsuya, Matsuda, Yoshiki, Higashino, Junichi
• Primary Examiner(s): Black, Thomas G.
• Assistant Examiner(s): Coby, Frantz
• Application Number:
  US08/536,258
• Time in Patent Office: 1,131 Days
• Field of Search: 395/601, 395/602, 395/700, 395/147, 395/603, 395/604, 382/54, 364/300, 364/401, 707/1, 707/2, 707/3, 707/4
• US Class Current: 1/1
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: Y10S 707/99931 Database or file accessing
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation

Abstract

A document information classification method and apparatus for classifying a document group and arranging a classified result hierarchically on the basis of key words given to the document group and words appearing in documents without dependence on a prescribed classification system. The document group of a document data base and a key word group given to each document of a key word data base are managed by a data management unit. A document classification unit classifies documents into folders on the basis of individual key words and stores them. The folders having similar document groups are integrated. Whether the integration is effective or not is judged upon integration. Whether the inside of the integrated folder and the inside of unintegrated folders can be classified in detail or not is judged and a hierarchical classification system is prepared. A classified result is produced in CRT by a classified result output unit to provide environment in which a user can read out the classified result.

Claims

1. A document information classification method for classifying each of a plurality of previously stored document information into predetermined classification, said method comprising:

a step of storing said plurality of document information and a plurality of sets of key words included in each document information of said plurality of document information;

a step of comparing the key words with said document information to classify said document information into a plurality of first key-word folders for each stored key word;

a step of comparing document information in said first key-word folders with document information in other key-word folders to combine said first and other key-word folders when a predetermined number of same document information is included in said first and other key-word folders thereby to prepare second key-word folders; and

a step of classifying said plurality of previously stored document information into said second key-word folders.

2. A document information classification method according to claim 1, comprising:

a step of comparing document information in said second key word folders with each other to combine said document information having a certain correspondence into third key word folders thereby to classify said plurality of document information.

3. A document information classification method according to claim 2, comprising:

a step of previously designating the number of said second key word folders, and

wherein said step of comparing said document information in said second key word folders with each other comprises repeating the combination until said designated number of second key word folders is reached.

4. A document information classification method according to claim 2, comprising:

a step of confirming whether said third key word folders are caused to be continued or not on the basis of contents of each document in said third key word folder.

5. A document information classification method according to claim 1, comprising:

a step of comparing other key words of document information in said second key-word folder with each other thereby to cluster said document information into a plurality of fourth key-word folders for one of said other key-words; and

a step of classifying said document information in said second key-word folder into said plurality of fourth key-word folders.

6. A document information classification method according to claim 5, comprising:

a step of previously defining an uppermost number of document information classified into said plurality of fourth key-word folders, and

wherein said step of classifying into said fourth key-word folder comprising classifying said document information is repeated until said previously defined number.

7. A document information classification method according to claim 1, comprising:

a step of extracting words appearing in said plurality of previously stored document information, and

a step of using said extracted words as said key-words.

8. A document information classification method according to claim 1, comprising:

a step of comparing documents in said first key-word folders with each other to combine said first key-word folders having the number of coincident document information exceeding a predetermined number with each other to prepare said second key-word folders.

9. A document information classification method according to claim 1, comprising:

a step of preparing said second key-word folders by using any one of an appearance frequency and an appearance position of said key-word in said document information.

10. A document information classification method according to claim 1, comprising:

a step of calculating a characteristic vector on importance of each document information in said second key-word folders and calculating an average vector of each of said calculated characteristic vectors, and

a step of redividing said document information having a difference between said calculated average vector and characteristic vector exceeding a predetermined value within said second key-word folders.

11. A document information classification method according to claim 1, comprising

a step of calculating a characteristic vector on importance of each document information in said second key word folders, and

a step of using said calculated characteristic vectors to judge whether said second key word folders are combined or not.

12. A document information classification apparatus including memory means for previously storing a plurality of document information and a key word group in each document information of said document information and classifying said document information, comprising:

first key word processing means for comparing key words of said plurality of document information to classify said plurality of document information into a plurality of first key-word folders for each key word stored in said memory means; and

second key word processing means for comparing document information in said plurality of first key-word folders with document information in other key-word folders to combine said first and other key-word folders when a predetermined number of same document information are included in said first and other key-word folders thereby to prepare second key-word folder;

whereby said plurality of previously stored document information is classified into said second key word folder.

13. A document information classification apparatus according to claim 12, comprising:

related key word combination means for composing document information in said second key-word folders with each other to combine said document information having a certain correspondence into third key-word folders thereby to classify said plurality of document information.

14. A document information classification apparatus according to claim 13, comprising:

confirmation means for confirming whether said third key-word folders are caused to be continued or not on the basis of contents of each document information within said third key-word folder.

15. A document information classification apparatus according to claim 12, comprising:

second single key word processing means for comparing other key words of document information within said key-word folder with each other to thereby cluster said document information into a plurality of fourth key-word folders for one of said other key words; and

wherein said document information within said second keyword folder is classified into said plurality of fourth key-word folder.

16. A document information classification apparatus according to claim 12, comprising:

extraction means for extracting words appearing in said plurality of previously stored document information; and

key word preparing means for using said extracted words as said key words.

17. A document information classification apparatus according to claim 12, wherein

said related key word folder processing means compares said document information within said first key-word folders with each other to combine said first key-word folders having the number of coincident document information exceeding a predetermined value with each other to prepare said second key word folders.

18. A document information classification apparatus according to claim 12, wherein said second key-word folders processing means uses any one of an appearance frequency and an appearance position of said key word in said document information to prepare said related key word folder.

19. A document information classification apparatus according to claim 12, comprising:

average vector calculation means for calculating a characteristic vector on importance of each document within said second key-word folders and calculating an average vector of each of said calculated characteristic vectors; and

redivision means for redividing said document information having a difference between said calculated average vector and characteristic vector exceeding a predetermined value within said second key word folders.

20. A document information classification apparatus according to claim 12, comprising:

characteristic vector calculation means for calculating a characteristic vector on importance of each document within said second key-word folders; and

combination judgment means using said calculated characteristic vector to judge whether said second key-word folders are combined or not.

22. A document information classification apparatus according to claim 12, wherein

said first key word processing means clusters document information including a string of characters selected by an operator into a first key word folder.

23. A document information classification apparatus according to claim 22, comprising:

a thesaurus for indicating a hierarchical structure of said string of characters; and

hierarchical relation construction means for constructing a hierarchical relation of said second key word folder by using said thesaurus.

24. A document information classification apparatus according to claim 22, comprising

means for defining similarity between second key word folders on the basis of document information included in said second key word folder;

and

means for selecting a second key word folder similar to a related key word folder selected by an operator in accordance with said similarity.

25. A document information classification apparatus according to claim 24, comprising:

means for removing document information belonging to and similar to the same second key word folder as document information selected by the operator from said second key word folders.

21. A document information classification apparatus including memory means for previously storing a plurality of document information and a key word group in each document information of said document information and classifying said document information, comprising:

first key-word processing means for comparing key words of plurality of said document information to classify said plurality of document information into a plurality of first key-word folders for each key word stored in said memory means;

second key-word processing means for comparing said document information in said plurality of first key-word folders with document information in other key-word folders to combine said first and other key-word folders when a predetermined number of same document information are included in said first and different key-word folders thereby to prepare a second key word folder; and

means for combining said second key word folders specified by an operator with each other;

whereby said plurality of previously stored document information is classified into said second key word folder.