Login| Sign Up| Help| Contact|

Patent Searching and Data

Document Type and Number:
Japanese Patent JP3220885
Kind Code:

PURPOSE: To provide a system capable of automatically extracting almost all key words by utilizing document data in plural fields and the appearance frequency of each word in each field.
CONSTITUTION: Each document data (a) are divided into words by a document data word dividing part 1, the divided words are stored in a word division table (b) and the appearance frequency of each word in each document data is detected by a word appearance frequency detecting part 2 and registered in a word appearance frequency table (c). The appearance frequency of each word in each field is found out based upon the table (c) and registered in a field-sorted appearance word totalizing table by a field-sorted word appearance frequency totalizing part 3, a key word in each field is extracted from the document data by means of a key word unnecessary word dictionary) 4 based upon the appearance frequency of the totalizing table, an unnecessay word which is not a key word in each field is extracted by an unnecessary word extracting part 5, and the extracted key word in each field and the unnecessary word are registered in a key word/ unnecessary word dictionary (d). At the time of allocating a key word to document data, key words and unnecessary words stored in the dictionary (d) are referred to, a word to be a keyword is extracted from words in the document data to which the key word is to be allocated and the extracted key word is allocated to the document data by a key word allocating part 7.

Tadahiro Kiyama
Hiroshi Tsuji
Application Number:
Publication Date:
October 22, 2001
Filing Date:
June 18, 1993
Export Citation:
Click for automatic bibliography generation   Help
International Classes:
G06F12/00; G06F17/30; (IPC1-7): G06F17/30
Domestic Patent References:
Other References:
Attorney, Agent or Firm:
Shigeru Sasaoka (1 person outside)