Automatic Extraction of Chinese-English Synonyms Based on a Three-Phase Approach

Abstract

In the paper, we propose a novel method to extract automatically the Chinese-English Synonyms based on a three-phase approach TPA) without the usage of dictionary. The extraction processes are divided into two categories: synonyms which occurred 1) just once, and 2) occurred more than once in the Chinese corpus. The procedure is based on the three-phase system, in which the people’s writing custom and statistical approaches are used to find out the correct synonyms. Our methods also can extract the unknown words in documents. The empirical results demonstrate the precision rates of outside tests reach 89.1%, which proves that the proposed methods of three-phase approach in the paper is feasible and effective.