The Automatic Question Answering System Based On Latent Semantic Analysis

With the rapid development and broad popularization of Internet, it's very convenient for people to get information from web. Meanwhile, the volatile increase of web information makes users very difficult to obtain exact and correct information quickly. Question Answering aims at providing more powerful information access tools to help users overcome the problem of information overloading. Compared to the development of the technologies of English Question Answering and the application of Question Answering System, there is a few institutes and universities to do researches on Chinese Question Answering. So, this article brings us a Question Answering System based on Latent Semantic Analysis (LSA). As a statistical model for natural language, Latent Semantic Analysis (LSA) is known as a method for knowledge acquisition, induction and representation. With the Using of the base theory of LSA to construct our Question Answering System, we will find the questions belong to the Question Answering Gather which have the same meaning with the new question asked by users. For investigation, first we build a Semantic Space for library, then, we construct a Question Answering Gather by manual, at last, we carry out the Question Answering System on the platform of Web Service. The system will co work with three independence parts name Web Server, Web Service Server and Compute Server. The Web Server is in charge of the work for communicating with users. The Compute Server will finish the main logic arithmetic for Semantic Matching. As the bridge between Web Server and Compute Server, the Web Service Server will transfer all the searched results to the Web Server. Any of the users can use the Question Answering System by browser anywhere. After finishing the test of Question Answering System, we found that the rate of nicety of the answers is satisfactory.