Design And Practice Of Automatic Question-answering In Automobile Field Based On Knowledge Graph

As one of the most important applications in the field of artificial intelligence,the intelligent question answering system has received more and more attention from scholars in recent years.Compared to traditional search engines(such as Baidu,Google,etc.)returning cumbersome and rough results,the results of intelligent question answering systems tend to be more precise and concise.At the same time,with the improvement of deep learning technology and the improvement of computing power,the intelligent question answering system is more and more mature,basically replacing the traditional manual customer service and other question and answer scenarios.At this time,a good question answering system is urgently needed in the automotive field where the market is clear,to help users solve the problems encountered in buying or driving,thereby improving the user experience when driving.This paper first expounds the basic theories such as neural network,cyclic neural network,knowledge map,semantic analysis and other deep learning and intelligent question and answer;then,for the vertical field of the automotive industry,explore how to obtain the corresponding structured and semi-structured data from the Internet.And converted into a specific triple structure in the knowledge map;then,based on the processed triple structure,how to build a knowledge map in a specific field;finally,in the construction of a good knowledge map,combined with natural language processing A common semantic analysis method to explore how to implement a smart question answering system.In the subsequent experimental tests,the system can have a good response ability to the conventional problem,the answer is more accurate and the response time is shorter.The validity of the method and the feasibility of the research are proved.This paper independently completes the above work and developed the entire system.However,the current system has a single function and is simple.This article is also continuing to advance,continue to deepen,and expand the exploration of related issues.