Question classification in chinese restricted-domain based on SVM and domain dictionary

Question classification is very important for question answering. This paper presents our research work on automatic question classification through support vector machine approaches. Unlike the classification using only bag-of-word features, we exploit the domain knowledge and question-specific stop words in our model, and also present how to enrich bag-of-word approach by implementing feature attributes to facilitate the question categorization. When tested on the questions in cooking domain, our approach reaches an accuracy up to 86.34%, which promisingly outperforms the result of the baseline.