IP Library Granted Patent US 9,262,402
Granted Patent B2
US 9,262,402 · App. 14/117,297 · Granted Feb 16, 2016

Device, method and program for assessing synonymous expressions

Inventors: Kenji Tateishi (Tokyo, JP); Kai Ishikawa (Tokyo, JP)
Assignee: NEC CORPORATION
G06F17/2795
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,262,402
App. No.
14/117,297
Granted
Feb 16, 2016
Kind
B2
Abstract

A synonymous expression assessment device includes: synonymy assessment means for receiving input of binary relations each of which includes a nominal and a predicate, and assessing whether or not the input binary relations are synonymous using a similarity between input nominals and a similarity between input predicates; and inter-predicate similarity computation means for, when computing the similarity between the input predicates based on a distribution of occurrence frequencies of nominals that are in binary relations to the input predicate in a document set, performing the computation using a distribution of only nominals that are used in the same type of concept as the input nominal.

Claims (24)

1. A synonymous expression assessment device comprising:

a hardware including a processor,

a synonymy assessment unit, implemented by the processor, for receiving input of binary relations each of which includes a nominal and a predicate, and assessing whether or not the input binary relations are synonymous using a similarity between input nominals and a similarity between input predicates; and

an inter-predicate similarity computation unit, implemented by the processor, for, when computing the similarity between the input predicates based on a distribution of occurrence frequencies of nominals that are in binary relations to the input predicate in a document set, performing the computation using a distribution of only nominals that are used in the same type of concept as the input nominal.

2. A synonymous expression assessment device comprising:

a hardware including a processor,

a synonymy assessment unit, implemented by the processor, for receiving input of binary relations each of which includes a nominal and a predicate, and assessing whether or not the input binary relations are synonymous using a similarity between input nominals and a similarity between input predicates;

an inter-predicate similarity computation unit, implemented by the processor, for, when computing the similarity between the input predicates based on a distribution of occurrence frequencies of nominals that are in binary relations to the input predicate in a document set, performing the computation using a distribution of only nominals that are used in the same type of concept as the input nominal; and

an inter-nominal similarity computation unit, implemented by the processor, for, when computing the similarity between the input nominals based on a distribution of occurrence frequencies of predicates that are in binary relations to the input nominal in the document set, performing the computation using a distribution of only predicates that are used in the same type of concept as the input predicate.

3. A synonymous expression assessment device comprising:

a hardware including a processor,

a synonymy assessment unit, implemented by the processor, for receiving input of binary relations each of which includes a nominal and a predicate, and assessing whether or not the input binary relations are synonymous using a similarity between input nominals and a similarity between input predicates;

a concept class storage unit for storing types of concept classes to which predicates or nominals belong;

an occurrence frequency correction unit, implemented by the processor, for determining, with reference to the types of the concept classes stored in the concept class storage unit, a degree to which a predicate or a nominal included in a document set is used in the same concept as the input predicate or the input nominal, and correcting an occurrence frequency of a binary relation included in the document set according to the determined degree;

an inter-predicate similarity computation unit, implemented by the processor, for setting, as a feature value of the input predicate, a distribution of occurrence frequencies or corrected occurrence frequencies of nominals that are in binary relations to the input predicate in the document set, and computing a degree of similarity between feature values of the input predicates; and

an inter-nominal similarity computation unit, implemented by the processor, for setting, as a feature value of the input nominal, a distribution of occurrence frequencies or corrected occurrence frequencies of predicates that are in binary relations to the input nominal in the document set, and computing a degree of similarity between feature values of the input nominals.

4. The synonymous expression assessment device according to claim 1 , wherein a nominal in each binary relation also includes a particle that is in a case relation to a predicate in the binary relation.

5. The synonymous expression assessment device according to claim 1 , wherein the synonymy assessment unit assesses the input binary relations as synonymous, in the case where the similarity between the input nominals and the similarity between the input predicates satisfy a predetermined condition.

6. A synonymous expression assessment method, implemented by a processor, comprising:

receiving input of binary relations each of which includes a nominal and a predicate, and assessing whether or not the input binary relations are synonymous using a similarity between input nominals and a similarity between input predicates; and

when computing the similarity between the input predicates based on a distribution of occurrence frequencies of nominals that are in binary relations to the input predicate in a document set, performing the computation using a distribution of only nominals that are used in the same type of concept as the input nominal.

7. A non-transitory computer readable information recording medium storing a synonymous expression assessment program for causing a computer to execute:

a synonymy assessment process of receiving input of binary relations each of which includes a nominal and a predicate, and assessing whether or not the input binary relations are synonymous using a similarity between input nominals and a similarity between input predicates; and

an inter-predicate similarity computation process of, when computing the similarity between the input predicates based on a distribution of occurrence frequencies of nominals that are in binary relations to the input predicate in a document set, performing the computation using a distribution of only nominals that are used in the same type of concept as the input nominal.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 21, 2024
From: NEC CORPORATION
To: NEC ASIA PACIFIC PTE LTD.
Reel/Frame 066867/0886 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 12, 2013
From: TATEISHI, KENJI; ISHIKAWA, KAI
To: NEC CORPORATION
Reel/Frame 031587/0342 →
Priority Claims (1)
JP 2011-105589 · May 10, 2011 · national
Continuity (1)
Related Publication 20140343922A1 · Nov 20, 2014