Generation of synonym list from existing thesaurus

Patent metadata

Field	Value
Publication number	US-9760630-B2
Application number	US-201514827051-A
Country	US
Kind code	B2
Filing date	Aug 14, 2015
Priority date	Aug 14, 2015
Publication date	Sep 12, 2017
Grant date	Sep 12, 2017

Abstract

A computer-implemented method, system, and computer program product for generating a synonym list from an existing thesaurus includes preparing a first feature vector from a natural language query and preparing a second feature vector from a result of the natural language query, determining, using a processor, whether a combination of a first feature from the first feature vector and a second feature from the second feature vector is included as a synonym pair in the existing thesaurus, and generating the synonym list by adding the combination to the synonym list when the determination is positive. The result of the natural language query may be identified by a user browsing action or by a positive feedback from a user. The method further includes reading a log which includes a single set or plural sets of the natural language query and the result of the natural language query.

First claim

What is claimed is:

1. A computer-implemented method for generating a synonym list from an existing thesaurus, the method comprising:
preparing a first feature vector from a natural language query and preparing a second feature vector from a result of the natural language query;
determining, using a processor, whether a combination of a first feature from the first feature vector and a second feature from the second feature vector is included as a synonym pair in the existing thesaurus; and
generating the synonym list by adding the combination to the synonym list when the determination is positive.

2. The method according to claim 1 , wherein the result of the natural language query is identified by a user browsing action or by a positive feedback from a user.

3. The method according to claim 1 , further comprising
reading a log which includes a single set or a plurality of sets of the natural language query and the result of the natural language query.

4. The method according to claim 1 , wherein the preparation of the first feature vector further comprises preparing, for each first feature, a first weight of the first feature and the preparation of the second feature vector further comprises preparing, for each second feature, a second weight of the second feature; and
the method further comprising determining whether the combination of the first feature and the second feature is added to the synonym list, using the first weight and the second weight.

5. The method according to claim 4 , further comprising changing the first weight of the first feature, the second weight of the second feature, or combination thereof.

6. The method according to claim 1 , wherein the synonym list is for a specific purpose.

7. The method according to claim 6 , wherein the synonym list for the specific purpose is a synonym list for a specific technical field or a synonym list for a specific user or group.

8. A system, comprising:
a processor; and
a memory storing a program, which, when executed on the processor, performs a method for generating a synonym list from a natural language query, the method comprising:
preparing a first feature vector from a natural language query and preparing a second feature vector from a result of the natural language query;
determining, using a processor, whether a combination of a first feature from the first feature vector and a second feature from the second feature vector is included as a synonym pair in the existing thesaurus; and
generating the synonym list by adding the combination to the synonym list when the determination is positive.

9. The system according to claim 8 , wherein the result of the natural language query is identified by a user browsing action or by a positive feedback from a user.

10. The system according to claim 8 , further comprising:
reading a log includes a single set or a plurality of sets of natural language query and the result of the natural language query.

11. The system according to claim 8 , wherein the preparation of the first feature vector further comprises preparing, for each first feature, a first weight of the first feature and the preparation of the second feature vector further comprises preparing, for each second feature, a second weight of the second feature; and
the method further comprising determining whether the combination of the first feature and the second feature is added to the synonym list, using the first weight and the second weight.

12. The system according to claim 11 , further comprising changing the first weight of the first feature, the second weight of the second feature, or combination thereof.

13. The system according to claim 8 , wherein the synonym list is for a specific purpose.

14. The system according to claim 13 , wherein the synonym list for the specific purpose is a synonym list for a specific technical field or a synonym list for a specific user or group.

15. A computer program product for generating a synonym list from a natural language query, the computer program product comprising a non-transitory computer usable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform a method comprising:
preparing a first feature vector from a natural language query and preparing a second feature vector from a result of the natural language query;
determining, using a processor, whether a combination of a first feature from the first feature vector and a second feature from the second feature vector is included as a synonym pair in the existing thesaurus; and
generating the synonym list by adding the combination to the synonym list when the determination is positive.

16. The computer program product according to claim 15 , wherein the result of the natural language query is identified by a user browsing action or by a positive feedback from a user.

17. The computer program product according to claim 15 , further comprising:
reading a log which includes a single set or a plurality of sets of the natural language query and the result of the natural language query.

18. The computer program product according to claim 15 , wherein the preparation of the first feature vector further comprises preparing, for each first feature, a first weight of the first feature and the preparation of the second feature vector further comprises preparing, for each second feature, a second weight of the second feature; and
the method further comprising determining whether the combination of the first feature and the second feature is added to the synonym list, using the first weight and the second weight.

19. The computer program product according to claim 18 , further comprising changing the first weight of the first feature, the second weight of the second feature, or combination thereof.

20. The computer program product according to claim 15 , wherein the synonym list is for a specific purpose.

Assignees

• IBM

Inventors

• Takase Toshiro
• Koseki Akira

Classifications

• G06F40/247

  Thesauruses; Synonyms · CPC title

• G06F16/374 Primary

  Thesaurus · CPC title

• G06F16/3347

  using vector based model · CPC title

• G06F16/335

  Filtering based on additional data, e.g. user or group profiles (filtering in web context G06F16/9535, G06F16/9536) · CPC title

• G06F17/3069

  Physics · mapped topic