A Research On The Approach Of Autogenerating Test Paper On The Basis Of Genetic Algorithm

With the development of the computer technology and artificial intelligence ,the research on the system of autogenerating test paper has gradually aroused great concern of more and more experts. Firstly,the thesis probed the theoretical foundation of the item bank construction, analyzed the testing standard of an examination paper, as well as the parameters of the test papers and the examination questions. Secondly, the thesis expounded design idea, overall function and the realizing method of the general system of autogenerating test paper. Then, in order to improve the successful rate and working efficiency, it put forward the consistent test method for the requirement of forming a test paper and the item bank structure before generating a test paper to find out the unreasonable requirement. Finally, on the basis of analyzing a large quantities of references home and abroad, the thesis made a survey on the autogenerating test paper algorithms, and created its mathemetics model, taking the knowledge points, degree of difficulty, types of questions, teaching requirement rate as the major controlling parameters. Characterized by its automatic researching ability, artificial intelligence searching technology and its high speed of convergence, this algorithm can be used to resolve the problems in traditional autogenerating test paper such as the over-long coding, the difficulty of calculating the adaptability function and the repeat of the knowledge points in the same paper, which hence, meets the demands of autogenerating test paper. It also brought forth a new structure of chromosome and the fitness function, designed the genetic operation. The thesis developed on autogenetating test paper. The result of the experiments shows that the method is reasonable and effective.