Construction of an “Analects of Confucius” knowledge base including pragmatics information

Abstract

SUMMARYIn this paper, we propose an approach to constructing a knowledge base for the Analects of Confucius, which aims to promote a correct understanding of this text. The content of the Analects has not been categorized by topic and it always contains broader meanings. Thus, it is necessary to create a framework that can be used to manage and build the knowledge base for the text. The construction of knowledge bases in the past has focused on the investigation of words and of surface meanings and explicitly communicated meanings of passages, but these cannot be used for study of deeper meanings or implicitly communicated meanings. The present paper sets up a categorization system for the Analects, which is used to create a knowledge base by using pragmatics information, based on the utterance interpretation method in pragmatics. A question answering system that makes use of the knowledge base on the Analects was used for assessment of the approach. The results showed that answers based on a combination of pragmatics information and modern text interpretation outperformed by 32.2% answers extracted only from the modern text.