Design And Implementation Of Computer Network Course Question Answering System Based On Knowledge Graph

In recent years,the question and answer system based on the knowledge graph has attracted much attention from researchers.Compared with the traditional search engine to obtain knowledge,the question and answer system can more accurately understand the meaning of the question entered by the user,and quickly match the answer from the knowledge base to the user.Traditional Q&A based on search engines can only return a large number of relevant web pages to users.Users need to filter the required information from many web pages,which is not only low in accuracy but also low in efficiency.The traditional faceto-face education model can no longer meet the needs of students and parents.The computer network course question and answer system based on knowledge graphs has appeared in people’s field of vision.It can be used by students and teachers at the same time.It has developed rapidly in teaching and learning and has a good reputation.effect.Using the question-and-answer system to learn not only solves some of the time and space problems of traditional face-to-face learning,it improves the learning efficiency of students,but also eases the workload of teachers.Therefore,the computer network course question answering system based on the knowledge graph is of great significance.The main work carried out in this paper has the following aspects:First:the construction of computer network knowledge graph.Based on the computer network course electronic textbook data,the data is marked,and the Internet-related computer network course data is supplemented,the different data is processed and integrated,and the knowledge graph for the computer network course question and answer system is constructed.Second:Research on entity recognition and relationship recognition algorithms.Using the model combination of BiLSTM+CNN can solve the problem of long-distance dependence in the text to a certain extent,and can take into account the local and global features in the text.Secondly,combine the conditional random field(CRF)to obtain the dependency relationship between entities and improve the entity Recognition effect:The introduction of attention mechanism in relation recognition can effectively improve the effect of semantic analysis in relation recognition.Third:Complete the development of a question-and-answer system for computer network courses based on knowledge graphs.This paper implements a computer network course question-and-answer system based on the knowledge graph.The test system runs stably and can meet the needs of course knowledge question-and-answer and independent learning.