Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Knowledge graph question and answer method and system based on neighbor interaction network

A knowledge map and network technology, applied in the direction of reasoning methods, neural learning methods, biological neural network models, etc., can solve problems such as poor retention of information, different attention of neighbors, and influence on the accuracy of answering questions, etc., to improve Accuracy, the effect of improving accuracy

Pending Publication Date: 2022-02-22
SHANDONG NORMAL UNIV
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing knowledge embedding methods cannot capture the rich hidden semantic information in the knowledge graph for encapsulation, which affects the accuracy of answering questions.
[0006] At the same time, the embedding method based on the graph structure can obtain neighbor information. Although the graph structure is conducive to the acquisition of hidden information in the knowledge graph and enhances the semantic representation of entity nodes in the knowledge graph, the attention of entities to neighbors is different. The weight given to neighbor entities will add a lot of unnecessary redundant information
In traditional graph attention networks, the same entity should play different roles in different triplets, and the information of the entity itself is not well preserved in the new representations obtained by these entities

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Knowledge graph question and answer method and system based on neighbor interaction network
  • Knowledge graph question and answer method and system based on neighbor interaction network
  • Knowledge graph question and answer method and system based on neighbor interaction network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] This embodiment discloses a knowledge graph question answering method based on a neighbor interaction network. As described in the background, since knowledge graphs in the real world are usually incomplete, there may be no link between two entities. The multi-hop question answering method brings challenges. The knowledge graph embedding method judges whether there is a relationship between two entities based on the semantic relationship between entities to deal with the problem of missing links. There is a lot of hidden information in the knowledge graph that cannot be obtained by existing embedding methods.

[0045] This application can perform multi-hop reasoning in incomplete knowledge graphs, apply the graph attention network to multi-hop question answering based on knowledge graphs, encapsulate neighbor information in entity representations, and capture hidden information contained in knowledge graphs, making a breakthrough The constraints of the range of selectab...

Embodiment 2

[0103] The implementation mode of this specification provides a knowledge map question answering system based on neighbor interaction network, which is realized through the following technical solutions:

[0104] The knowledge map semantic module is configured to: obtain the knowledge map, convert the knowledge map into an embedded representation of entities and relationships of the knowledge map to form a semantic space according to the knowledge map and the neighbor interaction network;

[0105] The question answer acquisition module is configured to: represent the question according to the question and the pre-trained language model to obtain the vector representation of the question; put the vector representation of the question into the semantic space to predict the answer entity to obtain the answer to the question.

Embodiment 3

[0107] The implementation mode of this specification provides a computer device, including a memory, a processor, and a computer program stored on the memory and operable on the processor. It is characterized in that, when the processor executes the program, it implements the Steps of a knowledge graph question answering method based on neighbor interaction network.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a knowledge graph question and answer method and system based on a neighbor interaction network. The method comprises the following steps: obtaining a knowledge graph, converting the knowledge graph into embedded expressions of entities and relationships of the knowledge graph according to the knowledge graph and a neighbor interaction network, and forming a semantic space; representing a question according to the question and a pre-training language model to obtain a vector representation of the question; putting the vector representation of the question into the semantic space to predict an answer entity to obtain an answer to the question.

Description

technical field [0001] The present disclosure relates to the technical field of knowledge graph question answering, in particular to a knowledge graph question answering method and system based on a neighbor interaction network. Background technique [0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art. [0003] A very important research task in natural language processing is the question answering system. Natural language question answering has always been a hot research issue, from only being able to answer questions about specific knowledge content in a single field to now being able to answer questions in multiple fields and conduct multiple rounds of communication , from methods based on semantic analysis and information extraction to the widespread use of deep learning methods. With the increasing application of computers, we can often see the application of natural la...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F16/332G06F16/36G06F40/205G06F40/35G06K9/62G06N3/04G06N3/08G06N5/04
CPCG06F16/3329G06F16/367G06F40/35G06F40/205G06N3/04G06N3/08G06N5/04G06F18/214
Inventor 徐连诚郭启萌刘培玉朱振方
Owner SHANDONG NORMAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com