Key points are not available for this paper at this time.
An automatic question answering method based on knowledge map is proposed and optimized. In order to improve the accuracy and efficiency of automatic answers, this paper adopts an automatic question answering framework based on knowledge map. Under this framework, this paper mainly studies how to integrate classification and subgraph matching and apply it to practical problems. This paper adopts a method that combines deep neural network with knowledge map. Then a method based on subgraph matching is proposed to realize automatic identification of questions. Advanced similarity measurement methods are introduced in the similarity comparison process to accurately measure the similarity between questions and alternative solutions. Finally, the validity of the proposed model is verified by experiments, and the future research direction is prospected. Through simulation experiments on public databases, the algorithm we proposed improves the answer accuracy by 10% compared with the baseline algorithm and shortens the response time by 20%.
Wu et al. (Sun,) studied this question.