Research And Design Of Intelligent Question Answering System For Basic Computer Course

With the development of Internet technology and the MOOC courses, the Intelligent Answering also has drown more attention. For traditional teaching, the students usually only solve related problems in the learning process in the classroom by teachers, and can not be solved in time, while the Intelligent Answering usually has no time limit, and it can solve user unsure questions timely.Intelligent Answering system typically includes comprehension questions, information retrieval and answer extraction and selection. comprehension questions which contains the Chinese question classification and information retrieval which contains the sentence similarity. For Intelligent Answering system, the quality of system is determined by the accuracy of the final returned results. Typically question classification and sentence similarity will directly affect the accuracy of the system, therefore, the study of question classification and sentence similarity algorithm become the subject of this thesis.Aiming at the above problems and considering the specific needs of Southwest Jiaotong University Computer-based teaching counseling, this thesis puts forward the solution of intelligent answering system. First of all, on the basis of the relevant theoretical and technical research of intelligent answering system, the author determines the functional requirements of the system. Secondly, the author analysis the existing Chinese word segmentation algorithm and uses the ICTCALS2013segmentation tool which increases the professional vocabulary words to meet system requirements. Then the author analysis the existing Chinese question classification algorithms and sentence similarity algorithm, improves the support vector machine algorithm which uses in Chinese question classification, and based on morphological pattern and improved edit-distance method, the author put forword to a new sentence similarity, which makes the results greatly improved. Finally, in order to meet the basic requirements of the uses, the author used JAVA EE technology to develop a set of Web-based B/S system, which bases on the Chinese question classification algorithm and sentence similarity algorithm.