For the purpose of sentence understanding,we are constructing a Chinese large-scale-corpus based on semantic dependency relations.

The paper presents domestic and alien researches on word senses disambiguation,and improves the algorithm of word sense disambiguation based on Navie-Bayes discriminance by adding large weights for characteristic words of polysemous words with syntactic dependency into Bayes discriminance formula,which amends the effect of word sense diambiguation.

This paper proposes a new feature selection method based on syntactic and semantic information,using the question word,the main verb of the question,the dependency structure,the main noun and the top hypernym of the noun as features for classification.

We haveexperimented with different feature sets:Bag of Words,Dependency Structure,WordNet Hypernyms and SelectedWordNet Hypernyms of a category.

The transformation rules from dependency relationship tree - bank is get and XML is used to managing the set of rules, including these operations: accession, deletion, modification.

The results show that the feature extraction method using SVM based on dependency relations can get high classification accuracy.

Chinese Chunk Dependency Analysis Based on Support Vector Machines;