Design And Implementation Of An Intelligent Question Answering System Based On Internet

Intelligent question answering is integrated application of natural language processing, artificial intelligence and information retrieval. It automatically gives immediate response to a question described by natural language, and is of great significance in the distance education. This paper focuses on some kernel techniques on Intelligent Question answering system. The main aspects are as follows:1. Chinese segmentation technique. Combining a variety of technological advantages, the bidirectional Maximum Matching Method is adopted for the problem of natural sentence segmentation; statistical methods for the difficulty of ambiguity, and taking greater frequency as the result; the unknown is added to the unregistered table, and solved by teachers later.2. Rapid location technique of answer library. According to the fixed structure characteristics of teaching contents, a three-tier index structure of the knowledge library is presented to rapidly locate answer and to improve the efficiency of the system.3. Systemic security. The different kinds of encrypt and access control techniques are analyzed. The SHA - 512 encryption algorithm and therole based access control is used to insure the security of the system.4. The system is designed and implemented successfully according to the theory mentioned above. An effective, intelligent and safe model is constructed by mutual cooperation techniques including the on-line real-time aspects and intelligent aspects. At last, the system is programmed in virtue of Microsoft's DotNet Technology.5. System accuracy and the real-time of the system are tested. The accurate matching rate of answer is 93.5% and average needed time of query is about 160ms, the results show that the system is real-time and more accurate.