System and method for enhancing neural sentence classification

Patent metadata

Field	Value
Publication number	US-11544946-B2
Application number	US-201916728019-A
Country	US
Kind code	B2
Filing date	Dec 27, 2019
Priority date	Dec 27, 2019
Publication date	Jan 3, 2023
Grant date	Jan 3, 2023

Abstract

A system and method is disclosed for classifying natural language sentences by employing external knowledge to assist in constructing a knowledge base of sentences with a target meaning. The disclosed system and method provide a general sentence classification framework applicable for a knowledge-oriented domain (e.g., domain-specific knowledge). The system and method may be implemented in an intelligent automotive aftermarket assistance tool to assist with the identification of sentences describing specific problems and solutions for car repairs. In addition to the domain adaptability, the system and method is language-independent and could be applicable to any natural written language.

First claim

What is claimed is:

1. A method comprising:
generating a sentence representation of a textual sentence;
retrieving entities and relations from a knowledge graph that are relevant to the textual sentence;
generating a representation of the relevant entities and relations; and
combining the sentence representation and the representation of the relevant entities and relations to produce an enhanced sentence representation.

2. The method of claim 1 , wherein the generated sentence representation is a low-dimensional vector representation.

3. The method of claim 2 , further comprising:
encoding the textual sentence, by a long short-term memory network, to produce the low-dimensional vector representation.

4. The method of claim 2 , further comprising:
encoding the textual sentence, by a convolutional neural network, to produce the low-dimensional vector representation.

5. The method of claim 1 , wherein the retrieving of the relevant entities and relations comprises performing a full-text search over documents created from the knowledge graph.

6. The method of claim 5 , wherein the performing of the full-text search comprises querying the documents with the textual sentence as a query.

7. The method of claim 1 , wherein the enhanced sentence representation is used to train a supervised classifier on a human-annotated dataset.

8. The method of claim 1 , wherein the knowledge graph is an external knowledge graph.

9. One or more non-transitory storage media comprising computer-executable instructions that, when executed, perform a method comprising:
generating a sentence representation of a textual sentence;
retrieving entities and relations from a knowledge graph that are relevant to the textual sentence;
generating a representation of the relevant entities and relations; and
combining the sentence representation and the representation of the relevant entities and relations to produce an enhanced sentence representation.

10. The one or more non-transitory storage media of claim 9 , wherein the generated sentence representation is a low-dimensional vector representation.

11. The one or more non-transitory storage media of claim 10 , wherein the method further comprises:
encoding the textual sentence, by a long short-term memory network, to produce the low-dimensional vector representation.

12. The one or more non-transitory storage media of claim 10 , wherein the method further comprises:
encoding the textual sentence, by a convolutional neural network, to produce the low-dimensional vector representation.

13. The one or more non-transitory storage media of claim 9 , wherein the knowledge graph is an external knowledge graph.

14. The one or more non-transitory storage media of claim 9 , wherein the retrieving of the relevant entities and relations comprises performing a full-text search over documents created from the knowledge graph.

15. The one or more non-transitory storage media of claim 14 , wherein the performing of the full-text search comprises querying the documents with the textual sentence as a query.

16. The one or more non-transitory storage media of claim 9 , wherein the enhanced sentence representation is used to train a supervised classifier on a human-annotated dataset.

17. A computer system, comprising:
a processor; and
one or more computer-readable media comprising computer-executable instructions that, when executed by the processor, cause the computer system to perform a method comprising:
generating a sentence representation of a textual sentence;
retrieving entities and relations from a knowledge graph that are relevant to the textual sentence;
generating a representation of the relevant entities and relations; and
combining the sentence representation and the representation of the relevant entities and relations to produce an enhanced sentence representation.

18. The computer system of claim 17 , wherein the generated sentence representation is a low-dimensional vector representation.

19. The computer system of claim 18 , wherein the method further comprises:
encoding the textual sentence, by a long short-term memory network, to produce the low-dimensional vector representation.

20. The computer system of claim 18 , wherein the method further comprises:
encoding the textual sentence, by a convolutional neural network, to produce the low-dimensional vector representation.

Assignees

• Bosch Gmbh Robert

Inventors

• Araki Jun
• Hoang Khanh Linh

Classifications

• G06F18/241

  relating to the classification model, e.g. parametric or non-parametric approaches · CPC title

• G06N5/02

  Knowledge representation; Symbolic representation · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06N5/022

  Knowledge engineering; Knowledge acquisition · CPC title

• G06F16/367

  Ontology · CPC title