The semantic collocations play important roles in parsing Chinese phrases. They are useful for both semantic disambiguation and structural disambiguation. In this paper, a representation of phrase semantic rules for Chinese is presented to formulate such knowledge, and a corpus-based method was proposed to acquire phrase semantic rules from a Chinese phrase corpus annotated with syntactic and semantic information. The method includes a metarule-guided algorithm for mining cross-level association rules to acquire phrase semantic rules and an optimization algorithm to filter these rules. The experiment results showed the effectiveness of the method. Disambiguation performance of these resulting rules was quiet well. © 2008 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Zheng, X. L., Zhou, C. L., Shi, X. D., Li, T. Q., & Chen, Y. D. (2008). Automatic acquisition of phrase semantic rule for Chinese. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4993 LNCS, pp. 543–548). https://doi.org/10.1007/978-3-540-68636-1_61
Mendeley helps you to discover research relevant for your work.