Criminal Detection of Mystery Novel by Reinforcement Learning

Abstract

In recent years, there is an interpretability problem that even experts cannot explain the reasoning process of machine learning. A contest featuring interpretability, “First Knowledge Graph Reasoning Challenge 2018.” was held in Tokyo. A previous study presented a method an expert system employs to evaluate the reasoning process. However, using natural language processing on original documents in the process is time-consuming and the output is different compared with the original documents. In this study, we explain the reasoning process by tracing the pass of knowledge graph with a policy gradient of reinforcement learning.