SOLVING THE CHINESE PHYSICAL PROBLEM BASED ON DEEP LEARNING AND KNOWLEDGE GRAPH

Abstract

In recent years, problem solving, automatic proof and human-like test-tasking have become a hot spot of research. This paper focus on the study of solving physical problem in Chinese. Based on the analysis of physical corpus, it is found that the physical problem are made up of n-tuples which contain concepts and relations between concepts, and the n-tuples can be expressed in the form of UP-graph (The graph of understanding problem), which is the semantic expression of physical problem. UP-graph is the base of problem solving which is generated by using physical knowledge graph (PKG). However, current knowledge graph is hard to be used in problem solving, because it cannot store methods for solving problem. So this paper presents a model of PKG which contains concepts and relations, in the model, concepts and relations are split into terms and unique IDs, and methods can be easily stored in the PKG as concepts. Based on the PKG, DKP-solving is proposed which is a novel approach for solving physical problem. The approach combines rules, statistical methods and knowledge reasoning effectively by integrating the deep learning and knowledge graph. The experimental results over the data set of real physical text indicate that DKP-solving is effective in physical problem solving.