Integrating heterogeneous thesauruses for Chinese synonyms

摘要

This paper proposes a hierarchical mapping method that integrates two heterogeneous thesauruses, CCD and Cilin. The proposed method takes the advantage of organizing structures of two thesauruses and brings a novel integrated resource for Chinese synonyms. A novel resource with 125,146 synonym sets that covers 148,327 words is got. Furthermore, the empirical result also shows that the performance of the proposed hierarchical mapping is superior when compared with direct mapping.