Study on question classification approach mixing multiple semantic characteristics together

This article proposes such a question classification approach that integrates multiple semantic features. It is aimed at these two questions in Chinese question classification models: inaccurate semantic information extraction and too slow processing speed caused by too high Eigenvector dimension. With the help of HowNet and the support vector machine and syntactic and semantic information of question sentences taken into consideration, this method picks up four classification features - the interrogative, the main sememe of key words, the named entity and the singular/plural form of nouns - to classify factual question sentences. Within the process of sememe extraction, the meaning disambiguation technology is added in. Algorithms to combine above characteristics and produce a better result are presented and justified. Experiment of this method has been done in the Chinese question classification set of the information retrieval laboratory of HarBin Institute of Technology, and results show that the method with multiple integrated semantic features is better than that with single feature.