Research on Unsupervised WSD Method Based on Sense Categories;

Based on the vector space model, the mapping relationship between word sense and sense categories is set up by using Machine Readable Dictionary (MRD) and sense categories thesaurus; and then use the knowledge of sense categories to learn disambiguate feature in corpus unsupervisedly; finally, the disambiguate featur.

This paper presents a method of using soft pattern to discover semantic classes.

Implementation of Automatic Switchover for Input Method Between Chinese and English in Visual FoxPro with User-defined Class;

Through metaphoricalization,the temporal words which serve as the temporal background originally change into the entities and participants of the events,and thus serve as subjects of different semantic types.

The characteristics lie in the aspects of semantic types, morpheme types and whether the compounds are used as independent sentences.

From their syntactic formation, there are some clear differences,not only between the composed componems Of whole construction, but also between the degrees of complication,the linear length and the semantic type in items of comparison,and the semantic function in items of conclusion.