Statistical method for extracting, and displaying keywords in forum/message board documents

• US 6,865,715 B2
• Filed: 03/26/1998
• Issued: 03/08/2005
• Est. Priority Date: 09/08/1997
• Status: Expired due to Fees

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Matsui, Kunio, Tsuda, Hiroshi, Uchino, Kanji
• Primary Examiner(s): Paula, Cesar B
• Application Number:
  US09/048,026
• Publication Number:
  US 20010018698A1
• Time in Patent Office: 2,539 Days
• Field of Search: 715/500, 715/501.1, 715/513, 715/526, 715/531, 707/7, 707/10, 707/501, 707/513
• US Class Current: 715/277
• CPC Class Codes: G06F 16/30 of unstructured textual data document management systems G06F16/93
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: Y10S 707/99937 Sorting

Abstract

A group of documents in a forum/message board are stored as a document group database in a predetermined server computer. A document group analysis device analyzes a group of documents contained in each set of documents corresponding to each message board in the document group database. A summation device generates a meta-index, a thread index, and an indexed file. A display device displays a group of documents in a display format of a keyword view, a thread view, or an author view using a meta-index and a thread index.

Claims

1. A relevant document display apparatus for displaying a group of documents containing cross-referenced message documents contributed to at least one of a forum and a message board established through a computer network, the cross-referenced message documents having a cross-referenced relationship in which a document contributed earlier is referenced by documents contributed afterwards, said apparatus comprising:

a document group analysis device to classify a plurality of documents forming a set of documents into at least one group of cross-referenced documents by determining for each document in the set, which of the documents is referenced, and for extracting information about classification as document group information;

a document group keyword extraction device to extract keywords contained in a body of a document forming part of each group of documents based on a statistical analysis by referring to the document group information; and

a document group keyword display device to display a title of each document in each group of documents together with the keywords extracted from the body of the document corresponding to the title by said document group keyword extraction device, the title being displayed in a title-displaying field and the keywords being displayed in a keyword-displaying field.

2. The apparatus according to claim 1, wherein said document group keyword display device displays with enhancement a group of documents containing a document specified by a user for retrieval, or a keyword specified for retrieval.

3. A relevant document display apparatus for displaying a group of documents containing cross-referenced message documents contributed to at least one of a forum and a message board established through a computer network, the cross-referenced message documents having a cross-referenced relationship in which a document contributed earlier is referenced by documents contributed afterwards, said apparatus comprising:

a document group analysis device to classify a plurality of documents forming a set of documents into at least one group of cross-referenced documents by determining for each document in the set, which of the documents is referenced, and for extracting information about classification as document group information;

a topic analysis device to further classify each group of cross-referenced documents based on topics extracted from each document forming part of each group of cross-referenced documents, and to extract information about further classification as topic classification information;

a topic keyword extraction device to extract keywords contained in a body of each document relevant to each topic obtained by the further classification of each of the groups of documents based on a statistical analysis by referring to the document group information and the topic classification information; and

a topic keyword display device to display a title of each document relevant to each topic obtained by the further classification together with the keywords extracted from the body of each document corresponding to each title by said topic keyword extraction device, the title being displayed in a title-displaying field and the keywords being displayed in a keyword-displaying field.

4. A relevant document display method for displaying a group of documents containing cross-referenced message documents contributed to at least one of a forum and a message board established through a computer network and having a cross-referenced relationship in which a document contributed earlier is referenced by documents contributed afterwards, said method comprising:

classifying a plurality of documents forming a set of documents into at least one group of cross-referenced documents by determining for each document in the set, which of the documents is referenced;

extracting information about classification as document group information;

extracting keywords contained in a body of a document forming part of each group of documents based on a statistical analysis by referring to the document group information; and

displaying a title of each document in each group of documents together with the keywords extracted from the body of the document corresponding to the title, the title being displayed in a title-displaying field and the keywords being displayed in a keyword-displaying field.

5. The method according to claim 4, further comprising displaying with enhancement a group of documents containing a document specified by a user for retrieval, or a keyword specified for retrieval.

6. A relevant document display method for displaying a group of documents containing cross-referenced message documents contributed to at least one of a forum and a message board established through a computer network and having a cross-referenced relationship in which a document contributed earlier is referenced by documents contributed afterwards, said method comprising:

classifying a plurality of documents forming a set of documents into at least one group of cross-referenced documents by determining for each document in the set, which of the documents is referenced;

extracting information about classification as document group information;

further classifying each of the classified group of documents based on topics extracted from each document forming part of each group of documents, and extracting information about further classification as topic classification information;

extracting keywords contained in a body of each document relevant to each topic obtained by the further classification of each group of cross-referenced documents based on a statistical analysis by referring to the document group information and the topic classification information; and

displaying a title of each document relevant to each topic obtained by the further classification together with the keywords extracted from the body of each document corresponding to each title, the title being displayed in a title-displaying field and the keywords being displayed in a keyword-displaying field.

7. A computer-readable storage medium storing instructions to direct a computer to perform a method for displaying a group of documents containing cross-referenced message documents contributed to at least one of a forum and a message board established through a computer network and having a cross-referenced relationship in which a document contributed earlier is referenced by documents contributed afterwards, said method comprising:

classifying a plurality of documents forming a set of documents into at least one group of cross-referenced documents by determining for each document in the set, which of the documents is referenced;

extracting information about classification as document group information;

extracting keywords contained in a body of a document forming part of each group of documents based on a statistical analysis by referring to the document group information; and

displaying a title relevant to each group of documents together with the corresponding keywords extracted from the bodies of the group of documents, the title being displayed in a title-displaying field and the keywords being displayed in a keyword-displaying field.

8. A relevant document display method for displaying a group of documents containing cross-referenced message documents contributed to at least one of a forum and a message board established through a computer network and having a cross-referenced relationship in which a document contributed earlier is referenced by documents contributed afterwards, said method comprising:

classifying a plurality of documents forming a set of documents into at least one group of cross-referenced documents by determining for each document in the set, which of the documents is referenced;

extracting information about classification as document group information;

automatically extracting keywords contained in a body of a document forming part of each group of documents based on a statistical analysis by referring to the document group information and without requiring user input; and

displaying a title of each document in each group of documents together with the keywords extracted from the body of the document corresponding to the title, the title being displayed in a title-displaying field and the keywords being displayed in a keyword-displaying field.