Design And Implementation Of Automatic Question Answering System Based On Movie Knowledge Graph

In today's life,the mass media has become an indispensable tool in people's audio-visual life.It affects people's perception of things in all aspects and areas,and then affects their thinking and changes their thinking and judgment of things.Among them,the movie is recognized as the most common and the most information-intensive medium.And because of people's pursuit of high quality of spiritual life,people's demand for movies is also increasing.At the same time,with the extensive coverage of the Internet,the amount of information related to movies is also increasing,so how to design and build a question and answer system about movies is becoming more and more urgent and important.The original purpose of the knowledge graph is to describe the various entities and concepts that exist objectively in the real world and the relationships between them.It can be regarded as a big picture.The knowledge graph scheme can understand the abstract parts of the language and confirm the information through interaction with the vague parts of the language better.It can lock more accurate results on recommendation and it will be easily achieved by algorithms.Because of these advantages,the question answering system on films in this paper is based on knowledge graph.This system aims to build an intelligent question answering system based on the movie knowledge graph.According to the software V model development process,first of all,it starts from the requirements,comprehensively analyzes the system from both functional and non-functional perspectives,and then expands the overall design,functional module and database design based on the requirements that have been analyzed.The system is divided into three parts from the overall function:(1)Data acquisition: use web crawler technology to crawl the data of movie wetsite and store it in the My SQL database.(2)Construction of knowledge graph: store data and the relationship between the data in the My SQL database in the Neo4 j graph database.(3)Question and answer processing: it is divided into three parts which include entity intention extraction,question and answer template matching,and answer sentence generation.Apply Chinese word segmentation,Naive Bayes classifier and Cypher sentence to match query graph database.The system is implemented in Python.Finally,a more comprehensive test of the system was launched.The automatic question answering system based on the film knowledge graph solves the current demand for a large number of film problems,provides a more intelligent and convenient way to feedback,and reduces the cost of manual responses and manual queries.