Supercharge Your Innovation With Domain-Expert AI Agents!

Intelligent Question Answering System Based on Knowledge Graph Subgraph Retrieval

A knowledge map and intelligent question answering technology, applied in the field of knowledge map, can solve problems such as complex problems and unsatisfactory results, and achieve the effects of improving the degree of relevance, improving the accuracy of recognition, and improving the accuracy of question and answer

Active Publication Date: 2022-04-01
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the actual question-and-answer scenario, the user's question is often more complex. Once the question is too long or there are multiple relationships and entities, the effect of the existing method is not satisfactory.

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
  • Intelligent Question Answering System Based on Knowledge Graph Subgraph Retrieval
  • Intelligent Question Answering System Based on Knowledge Graph Subgraph Retrieval
  • Intelligent Question Answering System Based on Knowledge Graph Subgraph Retrieval

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] This embodiment provides an intelligent question answering system based on knowledge map subgraph retrieval, including: a question sentence processing module, a knowledge map embedding module, a knowledge map subgraph retrieval module, a knowledge map subgraph filtering module, and an answer output module;

[0047] The question processing module identifies the entities in the user input questions, constructs a syntactic dependency tree of the input questions according to the entities, and obtains relational predicates between entities;

[0048] The knowledge map embedding module converts the entity and the relationship predicate data between entities obtained by the question processing module into a low-dimensional dense vector;

[0049] The knowledge graph embedding module represents entities, relationships, attributes or values ​​of the knowledge graph as low-dimensional dense vectors;

[0050] The knowledge map subgraph retrieval module maps the entities in the quest...

Embodiment 2

[0078] Such as figure 1 The intelligent question answering system based on knowledge map subgraph retrieval provided by the embodiment of the application shown includes:

[0079] Question processing module, knowledge map embedding module, knowledge map subgraph retrieval module, knowledge map subgraph filtering module and answer output module;

[0080] Assume that the length of the input question is n; the question processing module identifies the entity in the user input question, builds the syntactic dependency tree of the input question according to the entity, and obtains the relational predicate between the entities. The specific process includes:

[0081] Use the BERT model to train the user input question, and get the semantic representation vector of the question C =BERT( n ), C ={ C 1 , C 2 ,…, C n };

[0082] Then the semantic representation vector of the question C Input BiLSTM to get the hidden layer matrix;

[0083]

[0084] in and Respectively r...

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 present invention provides an intelligent question answering system based on knowledge map subgraph retrieval, including: the question processing module identifies the entities in the user input questions, constructs a syntactic dependency tree of the input questions according to the entities, and obtains the relationship between the entities Predicate; use the knowledge graph embedding to obtain the low-dimensional dense vector representation of triples in the knowledge graph; map the entities in the obtained user input questions to the knowledge graph, and obtain the candidate entity list corresponding to each entity in the knowledge graph, Select the starting entity, introduce the relational predicate, retrieve the matching knowledge map subgraph in the knowledge map, and then introduce the entity and relational predicate to continue the search, iterate until all the entities and relational predicates are introduced, and output the retrieved knowledge map subgraph; then use knowledge graph subgraph filtering to prune the retrieved knowledge graph subgraph; finally, output the corresponding results.

Description

technical field [0001] The invention relates to the field of knowledge graphs, in particular to an intelligent question answering system based on knowledge graph subgraph retrieval. Background technique [0002] With the development of science and technology, Internet applications have penetrated into all aspects of human life. Information retrieval is an important technical application under the background of the Internet and a necessary means for people to obtain information. At present, the main way for people to collect information is through traditional search engines. This search method generally only requires users to input keywords, and the search engine will adopt keyword matching methods to present all search results that meet the matching conditions to users. However, this method is difficult to meet the needs of users: on the one hand, the most direct answer cannot be returned to the user, but some web pages or documents related to the question, and the user need...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/332G06F16/36G06F40/295
CPCG06F16/3329G06F16/367G06F40/295
Inventor 陶建华于敏张大伟刘通杨国花
Owner INST OF AUTOMATION CHINESE ACAD OF SCI
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More