Research and Design of Online System Based on Subjective Questions Automatic Scoring

Abstract

The examinations of college public courses are great challenges for teachers who are responsible for large numbers of college students. With the traditional examination model, teachers need to set questions, set types, invigilate, correct, record scores and add up the score. The complex and various work costs them a lot of energy and spending. Online examination system is a system based on web technology that expands the flexibility of the examination and ensures the safety of the answers. However, at present, the questions in the online examination system are too single and they also need to be graded by teachers manually. Hence, this paper aims to develop an automatic scoring system for online subjective questions in the basic courses tests. Based on the research of the automatic grading technology at home and abroad, the online examination system realize automatically grading function of subjective questions, combined with Chinese word segmentation technology, keyword extraction algorithm, the statement in the matching processing algorithm, degree theory in fuzzy mathematics thought and algorithm analysis. This research simulates the artificial evaluation steps, and detailing the evaluation process step by step. According to the particularity of natural language, it establishes a synonym thesaurus for keyword matching in order to improve the accuracy of matching. Finally combing the close of the sentence, it calculates score by putting keywords as a scoring point, and the system is feasible through testing.