Keyword assigning method and system therefor

• US 5,642,518 A
• Filed: 06/13/1994
• Issued: 06/24/1997
• Est. Priority Date: 06/18/1993
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Kiyama, Tadahiro, Tsuji, Hiroshi
• Primary Examiner(s): Weinhardt, Robert A.
• Assistant Examiner(s): Shingala, Gita
• Application Number:
  US08/258,755
• Time in Patent Office: 1,107 Days
• Field of Search: 364/419.19, 364/419.08, 395/600, 395/757, 395/759, 395/760
• US Class Current: 704/7
• CPC Class Codes: G06F 16/345 Summarisation for human users
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A keyword assigning system divides text data into words, stores results of the division in a table, detects numbers of occurrence times of words in each text data, registers them in a table, and obtains word occurrence frequencies in domains on the basis of the table. The system then extract keywords from the text data by domain on the basis of a resulted domain-wise word occurrence collection table, and extracts negligible word not capable of becoming keywords by domain. The system further registers the extracted domain-wise keywords and negligible words in a keyword-negligible word dictionary, looks the keywords and negligible words registered in the dictionary upon assignment of the keywords to the text data to extract words to become keywords from among the words of the text data to have the keywords assigned thereto, and assign the extracted keywords to the text data, thereby automatically making extraction and assignment of most of the keywords.

Claims

1. A keyword assigning method in a text processing system having a memory and a processor, comprising:

a step performed by said processor, of extracting domain-wise keywords from text data in one of a plurality of texts, on the basis of occurrence frequencies of domain-wise word occurrence collection data stored in a memory;

a step of extracting negligible words, each occurring at a frequency of more than a specific occurrence ratio in all of said plurality of texts, by consulting said domain-wise word occurrence collection data;

a step of storing the extracted domain-wise keywords and the extracted domain-wise negligible words to said memory;

a step of inputting a text for assigning a keyword, to said text processing system;

a step of extracting a word as a keyword by way of consulting the stored keywords and the stored negligible words; and

,

a step of assigning the extracted keyword to the inputted text.

2. The keyword assigning method according to claim 1, wherein the step of extracting the domain-wise keywords from the text data extracts words appearing in only one domain and at the same time, not appearing in other domains as keywords in the domain having the words appeared therein.

3. The keyword assigning method according to claim 1, wherein the step of extracting the negligible words extracts words appearing in all said texts in all domains as negligible words common to all the domains.

4. The keyword assigning method according to claim 1, wherein the step of extracting the negligible words extracts words appearing in all said texts in one or more domains as negligible words in all the domains.

5. The keyword assigning method according to claim 1, wherein each of the steps of storing the domain-wise keywords and the domain-wise negligible words is a dictionary editing step of registering occurrence frequencies of words other than the ones extracted as the keywords and the negligible words, into a keyword-negligible word dictionary on the basis of the domain-wise word occurrence collection data.

6. The keyword assigning method according to claim 5, wherein the dictionary editing step further comprises:

a substep of, after displaying occurrence frequencies of the words other than the ones extracted as the keywords and the negligible words, by domain on the basis of the domain-wise word occurrence collection data, displaying numbers of text-wise occurrence times of words indicated by a user on the basis of the domain-wise word occurrence collection data.

7. The keyword assigning method according to claim 1, wherein the keyword assigning step further comprises:

a substep of dividing the text data of said inputted text into words by consulting a general dictionary for assigning the words as keywords, if there exist the words to become keywords in the text data in the domain thereof, by consulting the stored keywords and the stored negligible words.

8. The keyword assigning method according to claim 1, wherein the keyword assigning step further comprises:

a substep of dividing the text data of said inputted text into words by consulting a general dictionary;

a substep of assigning the words as keywords; and

a substep of consulting the stored keywords and the stored negligible words, and if there exists words which are not to become keywords nor negligible words in the text data in the domain thereof, assigning the words as keywords thereto.

9. The keyword assigning method according to claim 1, wherein the keyword assigning step further comprises:

a substep of dividing the text data of said inputted text into words by consulting a general dictionary; and

,

a substep of assigning words as keywords if the words are not to become keywords nor negligible words in the text data in the domain thereof by consulting the stored keywords and the stored negligible words and if the assignment is externally requested.

10. The keyword assigning method according to claim 8, wherein the keyword assigning step further comprises:

a substep of registering words not to become keywords nor negligible words in the domain for the text data into said memory.

11. A keyword assigning method in a text processing system having a memory and a processor, comprising:

a step of detecting numbers of occurrence times of words in one of a plurality of text data;

a step of storing the numbers of occurrence times as word occurrence frequency data;

a step of obtaining occurrence frequencies of words in each of domains on the basis of the word occurrence frequency data;

a step of storing the obtained occurrence frequencies as domain-wise word occurrence collection data;

a step of extracting domain-wise keywords from said text data on the basis of occurrence frequencies of the stored domain-wise word occurrence collection data;

a step of extracting negligible words, each occurring at a frequency of more than a specific occurrence ratio in all of said plurality of texts, by consulting said domain-wise word occurrence collection data;

a step of storing the extracted domain-wise keywords and the extracted domain-wise negligible words to said memory;

a step of inputting a text for assigning a keyword, to said text processing system;

a step of extracting a word as a keyword by way of consulting the stored keywords and the stored negligible words in said memory; and

,

a step of assigning the extracted keyword to the text data.

12. The keyword assigning method according to claim 11, wherein the step of detecting the numbers of occurrence times of the words in one of the plurality of the text data further comprises:

a substep of detecting items composing the text data, detecting numbers of occurrence times of words by item in the text data, and obtaining total numbers of occurrence times of the words in the text data.

13. The keyword assigning method according to claim 11, wherein the step of obtaining occurrence frequencies of the words in each of the domains further comprises:

a substep of obtaining a ratio of a number of text data in which a word occurs, to a number of text data in a single domain, as occurrence frequency of the word.

14. The keyword assigning method according to claim 11, wherein the step of storing comprises generating a keyword-negligible word dictionary for a plurality of domains.

15. The keyword assigning a method according to claim 11, wherein the keyword assigning step further comprises:

a substep of dividing the text data of said inputted text into words by consulting a general dictionary;

a substep of assigning the words as keywords;

a substep of consulting the stored keywords and the stored negligible words, and if there exists words which are not to become keywords nor negligible words in the text data in the domain thereof, assigning the words as keywords thereto; and

,

a substep of registering words not to become keywords nor negligible words in the domain for the text data into said memory.

16. A keyword assigning system in a text processing system having a memory and a processor, comprising:

means for extracting domain-wise keywords from text data in one of a plurality of texts, on the basis of occurrence frequencies of domain-wise word occurrence collection data stored in a memory;

means for extracting negligible words, each occurring at a frequency of more than a specific occurrence ratio in all of said plurality of texts, by consulting said domain-wise word occurrence collection data;

means for storing the extracted domain-wise keywords and the extracted domain-wise negligible words to said memory;

means for inputting a text for assigning a keyword, to said text processing system;

means for extracting a word as a keyword by way of consulting the stored keywords and the stored negligible words in said memory; and

,

means for assigning the extracted keyword to the inputted text.

17. The keyword assigning system according to claim 16, wherein the step of storing comprises generating a keyword-negligible word dictionary for a plurality of domains.

18. The keyword assigning system according to claim 16, wherein the keyword assigning step further comprises:

a substep of dividing the text data of said inputted text into words by consulting a general dictionary;

a substep of assigning the words as keywords;

a substep of consulting the stored keywords and the stored negligible words, and if there exists words which are not to become keywords nor negligible words in the text data in the domain thereof, assigning the words as keywords thereto; and

,

a substep of registering words not to become keywords nor negligible words in the domain for the text data into said memory.

19. A keyword assigning system in a text processing system having a memory and a processor, comprising:

means for detecting numbers of occurrence times of words in one of a plurality of text data;

means for storing the numbers of occurrence times as word occurrence frequency data;

means for obtaining occurrence frequencies of words in each of domains on the basis of the word occurrence frequency data;

means for storing the obtained occurrence frequencies as domain-wise word occurrence collection data;

means for extracting domain-wise keywords from said text data on the basis of occurrence frequencies of is the stored domain-wise word occurrence collection data;

means for extracting negligible words, each occurring at a frequency of more than a specific occurrence ratio in all of said plurality of texts, by consulting said domain-wise word occurrence collection data;

means for storing the extracted domain-wise keywords and the extracted domain-wise negligible words to said memory;

means for inputting a text for assigning a keyword, to said text processing system;

means for extracting a word as a keyword by way of consulting the stored keywords and the stored negligible words in said memory; and

,

means for assigning the extracted keyword to the text data.

20. The keyword assigning a system according to claim 19, wherein the keyword assigning step further comprises:

a substep of dividing the text data of said inputted text into words by consulting a general dictionary;

a substep of assigning the words as keywords;

a substep of consulting the stored keywords and the stored negligible words, and if there exists words which are not to become keywords nor negligible words in the text data in the domain thereof, assigning the words as keywords thereto; and

,

a substep of registering words not to become keywords nor negligible words in the domain for the text data into said memory.