Research on Design of Automatic Question Answering System Based on Convolutional Neural Network

摘要

Question answering system is an emerging information service system, which includes technologies such as natural language processing, semantic analysis, information retrieval, and artificial intelligence. Question answering system is currently a research hotspot in the field of natural language processing. It not only allows users to directly ask questions through natural language, but also directly returns accurate and concise answers to users. The question and answer system is different from the search engine that adopts question and answer technology. It can not only use natural language for question and answer, but also directly return accurate answers to users, and can answer users' further questions, which greatly improves user satisfaction and can better meet the real needs of users. This paper uses word convolutional neural network to build a set of high-accuracy task-oriented question answering system, improves the existing convolutional neural network question classification algorithm, and compares experiments on commonly used data sets, and compares experimental results. It highlights the validity and rationality of the question and answer algorithm designed in this paper, and makes a more detailed analysis of the experimental results.