Knowledge analysis system, knowledge analysis method, and knowledge analysis program product

• US 20030023600A1
• Filed: 02/27/2002
• Published: 01/30/2003
• Est. Priority Date: 07/30/2001
• Status: Active Grant

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha
• Inventors: Sasaki, Junya, Shimakawa, Kazunori, Kondo, Nobuaki, Atsumi, Kazuhiko, Nagamura, Eiji, Takachio, Katsuhiko
• Granted Patent: US 6,895,397 B2
• Time in Patent Office: Days
• US Class Current: 707/10
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: G06F 16/358 Browsing; Visualisation therefor

Abstract

There is provided a knowledge analysis system configured to be connectable to plural client terminals via network, which supports analysis requested by each of client terminals to knowledge accumulated in knowledge database, comprising: access control means for conducting user authentication to the client terminal requesting for access for permitting knowledge analysis from the client terminal; and knowledge analysis means for clustering knowledge accumulated in the knowledge database to create cluster database in which each of the knowledge is classified into clusters defined based on category; wherein the knowledge analysis means has means for setting important words having priority in clustering at creation of the cluster database, and carries out clustering to create an axis of cluster on the basis of the important words.

Claims

1. A knowledge analysis system configured to be connectable to plural client terminals via network, which supports analysis requested by each of client terminals to knowledge accumulated in knowledge database, comprising:

access control means for conducting user authentication to the client terminal requesting for access for permitting knowledge analysis from the client terminal; and

knowledge analysis means for clustering knowledge accumulated in the knowledge database to create cluster database in which each of the knowledge is classified into clusters defined based on category;

wherein

the knowledge analysis means has means for setting important words having priority in clustering at creation of the cluster database, and carries out clustering to create an axis of cluster on the basis of the important words.

2. A knowledge analysis system according to claim 1, wherein the knowledge analysis means has means for setting unnecessary words to be ignored in clustering, and carries out clustering on the basis of the important words and the unnecessary words.

3. A knowledge analysis system according to claim 2, wherein the knowledge analysis means has means for setting synonymous words to be handled as identical words in clustering, and carries out clustering on the basis of the combination of the important words and the unnecessary words and the synonymous words.

4. A knowledge analysis system according to claim 1, wherein the knowledge analysis means has means for storing analysis conditions used at creation of the cluster database.

5. A knowledge analysis system according to claim 4, wherein the knowledge analysis means has means for reading the stored analysis conditions, and creating new cluster database by use of analysis conditions which are changed from the stored analysis conditions.

6. A knowledge analysis system according to claim 1, wherein the knowledge analysis means has means for creating cluster database from knowledge accumulated in the knowledge data on the basis of analysis conditions including important words, unnecessary words and synonymous words;

and

re-analysis means for obtaining the analysis conditions, and carry out clustering once again by use of analysis conditions in which at least one or more of the important words, the unnecessary words and the synonymous words are reset to recreate cluster database and to replace the already-created cluster database.

7. A knowledge analysis system according to claim 6, wherein the re-analysis means further comprises cluster setting means for prompting the client terminal to set clusters to be objectives of recreation among the clusters contained in the already-created cluster database, and carrying out the re-clustering on clusters set by the client terminal.

8. A knowledge analysis system according to claim 1, wherein in the clustering, the knowledge analysis means determines a hierarchical structure defining hierarchical relation of one knowledge piece and another knowledge piece among the knowledge, and also determines clusters to which the respective knowledge pieces belong.

9. A knowledge analysis system according to claim 1, wherein the knowledge analysis means prompts a user to input clustering conditions including at least one of analysis result name, analysis objective period, focusing keyword, number of focused cases, number of hierarchies of hierarchical structure defining hierarchical relation of one knowledge piece and another knowledge piece among the knowledge, presence or absence of redundancy of knowledge, and number of most significant clusters to carry out clustering on the basis of the input clustering conditions.

10. A knowledge analysis system according to claim 1, wherein the knowledge analysis means further comprises editing processing means for editing the already-created cluster database and making the client terminal display the edited cluster database, and

the editing processing means prompts the client terminal to input editing conditions including at least one of presence or absence of cluster list display, presence or absence of time series display, presence or absence of hierarchical structure display, and presence or absence of graph display, and edits the client database on the basis of the editing conditions input by the client terminal, and makes the client terminal display editing processing results including at least one of cluster list display, time series display, hierarchical structure display, and graph display.

11. A knowledge analysis method for supporting analysis requested from each of client terminals to knowledge accumulated in knowledge database, comprising:

conducting user authentication to client terminals requesting for access for permitting knowledge analysis from the client terminals; and

clustering knowledge accumulated in the knowledge database to create cluster database in which each of the knowledge is classified into clusters defined based on category;

wherein in the creation of the cluster database, important words having priority in clustering are set to create an axis of cluster on the basis of the important words.

12. A knowledge analysis method according to claim 11, wherein in the creation of the cluster database, unnecessary words to be ignored in clustering are set, and the clustering is carried out on the basis of the important words and the unnecessary words.

13. A knowledge analysis method according to claim 12, wherein in the creation of the cluster database, synonymous words to be handled as identical words in clustering are set, and the clustering is carried out on the basis of the combination of the important words, the unnecessary words and the synonymous words.

14. A knowledge analysis method according to claim 11, further comprising:

storing analysis conditions used in the creation of the cluster database.

15. A knowledge analysis method according to claim 11, further comprising:

at creation of the cluster database, creating the cluster database from knowledge accumulated in the knowledge data on the basis of analysis conditions including important words, unnecessary words and synonymous words, and

obtaining the analysis conditions, and carrying out clustering once again by use of analysis conditions in which at least one or more of the important words and the unnecessary words and the synonymous words are reset to recreate cluster database and to replace the already-created cluster database.

16. A knowledge analysis method according to claim 11, wherein in the clustering, a hierarchical structure defining hierarchical relation of one knowledge piece and other knowledge piece among the knowledge is determined, and clusters to which the respective knowledge pieces belong are determined.

17. A knowledge analysis method according to claim 11, further comprising:

prompting the client terminal to input editing conditions including at least one of presence or absence of cluster list display, presence or absence of time series display, presence or absence of hierarchical structure display, and presence or absence of graph display;

editing the already-created client database on the basis of the editing conditions input by the client terminal to make the client terminal display editing processing results including at least one of cluster list display, time series display, hierarchical structure display, and graph display.

18. A knowledge analysis program product which supports for a computer system to analyze knowledge accumulated in knowledge database, comprising:

a recording medium;

a first program code which is recorded in the recording medium to assign the computer system a command to carry out user authentication to client terminals asking for access for permitting knowledge analysis from client terminals;

a second program code which is recorded in the recording medium to assign the computer system a command to create cluster database used for knowledge analysis from each of client terminals whose access is permitted, for classifying each of knowledge accumulated in the knowledge database into clusters defined based on category; and

a third program code which is recorded in the recording medium to assign the computer system a command to carry out analysis condition setting procedures to set important words having priority in clustering, unnecessary words to be ignored in clustering, and synonyms to be handled as synonymous words in clustering, at creation of the cluster database.

19. A knowledge analysis program product according to claim 18, further comprising:

a fourth program code which is recorded in the recording medium to assign the computer system a command to carry out analysis condition saving procedures to save the analysis conditions used at creation of the cluster database.

20. A knowledge analysis program product according to claim 19, further comprising:

a fifth program code which is recorded in the recording medium to assign the computer system a command to create cluster databases from knowledge accumulated in the knowledge data on the basis of analysis conditions of important words and unnecessary words and synonymous words, and

a sixth program code which is recorded in the recording medium to assign the computer system a command to obtain the analysis conditions used at creation of the cluster database and to re-create cluster database by use of analysis conditions after at least one or more of the important words and the unnecessary words and the synonymous words are reset and replace the already-created cluster database.

21. A knowledge analysis system configured to be connectable to plural client terminals via network, which supports analysis requested from each of client terminals to knowledge accumulated in knowledge database, comprising:

an access control device which conducts user authentication to the client terminal requesting for access for permitting knowledge analysis from the client terminals; and

a knowledge analysis device which carries out clustering to knowledge accumulated in the knowledge database to create cluster database in which each of the knowledge is classified into clusters defined based on category;

wherein

the knowledge analysis device has a clustering device which sets important words having priority in clustering in the creation of the cluster database, and carries out clustering to create an axis of cluster on the basis of the important words.