The construction and maintenance of the frequently asked question

With the development of Question Answering System which is a hot topic in the area of Artificial Intelligence, the set of FAQ (Frequently Asked Question) has become a widely used knowledge base and it do play an important role in Question Answering System. Besides this, it is also important in Information Retrieval. Therefore, the problem of how to construct and maintain the FAQ automatically has attracted many researchers ´ attention. In this paper, we put forward a method aims at solving the problem. We quote the “Concept Hierarchy Structure” for the first time which combined with the theory of text classification. We implement a method of computing similarity between sentences named TFIDF in the construction and maintenance of the FAQ. We do some experiments to verify our method and analyze the shortcomings of the method. From the experiment, we can see that the method introduced in this paper can reach a nice result.