A Classification of Questions Using SVM and Semantic Similarity Analysis

Abstract

Question classification is an important part in the question answering system. The results of the question classification determine the quality of the question answering system. In this paper, a question classification algorithm based on SVM and question semantic similarity is proposed, it is applied in a real-world on-line interactive question answering system in tourism domain. In the two level question classification method, Support Vector Machine model is adopted to train a classifier on coarse categories, question semantic similarity model is used to classify the question into sub-categories. The use of concept of domain terms construction will improve the feature expression of Support Vector Machine and question semantic similarity. The experimental result show that the accuracy of the classification algorithm is up to 91.49%.