The Design of an Intelligent Question Answering System on Restricted Domain

A intelligent question answering system model was presented in this paper, The system combined with semantics and vector space model to Automatically calculate the Semantic similarity of questions the user to enter and questions of the candidate question set. The candidate question set was built up according to the user's query by website crawler and artificially amending. Then, semantic edit and expansion were put up to users query via domain ontology. Finally the questions were found in the candidate question set and the corresponding answers are returned to users, which need the semantic similarity between the users query and the candidate question was up to the given threshold. Besides this, a catching mechanism was applied to improve the efficiency of the system; the system also maintained and updated the candidate question set by the feedback information of users. Experiments show that the system has better performance both in efficacy and satisfaction of user.