Font Size: a A A

Research On The Key Technology Of Automatic Question Answering System In Housing Fund Area

Posted on:2019-02-24Degree:MasterType:Thesis
Country:ChinaCandidate:J TaoFull Text:PDF
GTID:2428330548994883Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
Housing fund plays an extremely important role in people's life.At present,there is an increasing demand for knowledge counseling in the housing fund area.However,the state has still relatively little investment in the construction of knowledge automatic question answering system in the housing fund area,and the current counseling service still stays in the announcement,publishing books and community Q&A stage,far from reaching the level of intelligence.Therefore,based on the actual demand,this paper studies the key technologies of automatic question answering system in the area of housing fund.Based on the research results of this paper,this paper realizes this system and tries to improve the user experience while solving practical problems.The main contents of this paper are as follows:First of all,through the analysis of knowledge corpus in the field of housing fund,this paper designs the expansion of thesaurus algorithm,which is used to extract the basic vocabulary of domain from the corpus;Then,research and extract the relationship between entities and entities in the housing fund area,and build a knowledge graph of the housing fund area accordingly;Then,in-depth study of the question analytic technology,this paper put forward the domain question type identification algorithm that combines syntactic analysis and SVM model;Then,put forward the key word extraction algorithm based on domain knowledge graph and dependency syntactic analysis,which is used to extract the important information that users are interested in from the query;Then,construct the Cypher query template and combine the question type identification and keyword extraction results automatically generate executable Cypher query statements for retrieval of questions based on Neo4j graphics database answer;Then,design question matching algorithm for answering from the library for answers;Then,based on the above research,coding to achieve a housing fund automatic question answering system.Finally,the performance evaluation of the designed algorithm and the implemented system proves the feasibility of the proposed scheme in the construction of automatic question answering system in the field of housing fund.
Keywords/Search Tags:Housing Fund, Automatic Question Answering System, Knowledge Graph, Question Type Recognition, Question Keywords Extraction
PDF Full Text Request
Related items