Research On Intelligent Test Paper Composition System Based On Genetic Arithmetic

With the development of the computer technology and artificial intelligence, Computer-Aided Instruction on the production and development in the advancement of education give an enormous impetus to teaching reform and development of technology. With computer application the intelligent systems for producing test paper has been paid a lot of attention by many scientists. The research not only involves how to establish themathematical model of exam paper generating system, but also relates to the appliedalgorithms. Several algorithms have been applied in test paper auto-generation, such as prioritystrategy, randomization, and trace and trial strategy, genetic algorithm and so on.. However, each method has its advantages and disadvantages, so far there is not a good solution to the algorithm of the automatic test paper program.After analyzing many domestic and international academic papers, this paper analyzes exam paper generating problem and trys to use the GA to help solve this problem. To establish the automatic test paper mathematical model, taking the knowledge points, degree of difficulty, types of questions,teaching requirement rate as the major controlling parameters. Traditional algorithms of composing test paper have the disadvantages of slow convergence, lowsuccess rate and quality. This paper mainly studies the use of GA in the Combinationoptimization. In order to avoid slow convergence and local convergence of simple genetic algorithm (SGA)for intelligent test paper generation, a kind of improved genetic algorithm(IGA) has been proposed in this paper. The algorithm used is not the worst individual to participate in genetic strategies to maintain optimal strategy, as well as the dynamic adjustment of genetic operator and mutation operator, to expand the search space, the papers for global optimization. Improved genetic algorithmsegment real number code, putting the question of the same type on the sameadopts section,and then the question number maps gene directly. Real number code avoids decodingprocedure, it may enhance operation efficiently.The main work of this paper is making the research of how to set up an appropriate mathematical model for exam paper generating problem and how to use GA to solve this mathematical model. The first chapter is the exordium, introducing the academic background about this problem; The second chapter introduced the theory of technology that is the basis of GA. In the third chapter, paper on the main issues in the course of the study, detailed analysis of the constraints of the test paper, the application of mathematical methods for the test paper issues a mathematical model; In the forth chapter, focuses on improved genetic algorithm in the intelligent generating test paper; In the fifth chapter , design and implementation of the intelligent generating test paper system; The last chapter is a conclusion of the paper.