Semantic Analysis Model of Chinese MMT Based on Hierarchical Network of Concepts in Chinese–English Machine Translation

Abstract

Machine translation (MT) is a subfield of computational linguistics, which needs the technologies of natural language understanding (NLU), natural language processing (NLP), and the methods of artificial intelligence (AI). Deep understanding of Chinese helps to meet the semantic constraints of Chinese. Since linear and hierarchical structures of language are presented meantime, this paper proposes a semantic analysis model of Chinese multiple-branched and multiple-labeled tree (MMT) based on the hierarchical network of concepts (HNC) in Chinese–English machine translation, which executes the semantic analysis of Chinese sentence and generates HNC-MMT, and performs the conversion of the HNC-MMT to generate the translation. The word knowledge-base and rule-base are used to parse the semantic structure of Chinese sentence. Experiments show one of the most important tasks of semantic analysis of Chinese sentence, the global eigen-chunk recognition, achieves an accuracy above 85 %, which indicates the effectiveness of the model and the parsing method.