Design And Implementation Of Automatic Question

With the rapid development of Internet and information technology, the demand for programmers is increasing, so program design courses have attracted more and more attention of universities and students. But in the teaching process of traditional education and distance education, the question and answer session in the programming experiment courses remains problematic. The students' questions cannot be solved timely and effectively, which has a significant effect on students' learning efficiency and enthusiasm. There are two reasons accounting for this problem. From the perspective of students, Students as beginners, they lack extensive knowledge of grammar and debugging experience, therefore they can't accurately understand the compiler errors. And it's difficult for them to handle programming problems independently. From the perspective of teachers, in the process of classroom teaching, the class time and teachers' energy is limited, therefore teachers can't guide each student to solve problems. According to this reality, this paper is dedicated to research the automatic question-answering system of the programming experiment courses in order to optimize the defective situation of the question and answer session.Based on the theory of knowledge engineering and the rule engine technology, this paper launches the research on the automatic question-answering system. To realize the automatic question-answering system, there are three steps. The first step is to build a knowledge base of high quality and efficiency, which is the research focus of this thesis. Firstly, This article designs knowledge base from three aspects:knowledge representation, knowledge structure and knowledge reasoning mechanism. Secondly, taking "C language program design" experimental course as an example and Using Drools build a knowledge base in accordance to the following procedures:knowledge acquisition, knowledge representation, knowledge check, collision detection. The second step is to realize knowledge reasoning on the basis of the knowledge base. The paper codes to make knowledge base working and implements three types of knowledge reasoning, problem solving, Knowledge reasoning, intelligent navigation. The third step is to use Eclipse and other development tools to achieve hierarchically automatic question answering system of C programming experimental course in accordance to the overall structure of the system.On the basis of theoretical and technical research, this paper designs experiments to evaluate the case knowledge base and the automatic question-answering system. The experimental data shows that on the premise that you acquire sufficient knowledge, the recall rate, the accuracy rate; and the reasoning efficiency of the case knowledge base and of the question-answering system are comparatively high. Students' satisfaction with, the whole question-answering system is high, and students' satisfaction with recall rate and accuracy rate of the system is higher than the actual level. The experimental data strongly proves the validity of construction technology of the knowledge base and the feasibility of the automatic question answering system. In fact, besides the "C language program design" course, the construction technology of the knowledge base is also applicable to other programming experiment courses and other knowledge areas with clear causal knowledge. For example, the field of medical disease diagnosis, the network fault diagnosis domain, the field of crops diseases and pests diagnosis and prevention, etc.