Research On Intelligent Test Paper Generating Method Based On Genetic Algorithm

With the continuous development of computer application technology,the computer aided education system combined computer with education develops rapidly.Under the current quality education,the examination is still one of the main measurement standards in teaching ability and student achievement.At the same time,the standards of examinations with different levels for test paper are different.So,how to generate a reasonable and high quality paper rely on the algorithm is one of means to assess the combination of computer and education.Therefore,the study of efficient intelligent test paper is of practical value.In this paper,the intelligent method of genetic algorithm is deeply researched and analyzed.The problem of intelligent test is essentially a combinatorial optimization problem with multi-constrained and multi-objective,and the genetic algorithm has the ability of global search,which has strong advantages on such problems.However,the genetic algorithm is easy to fall into early convergence and converges slowly in late stage,so an improved genetic algorithm(NCAGA)is proposed in this paper.Based on the analysis of the convergence index of early convergence,NCAGA algorithm improves the crossover and mutation probability function,adjusts the individual according to the individual fitness to ensure the population diversity.At the same time,the algorithm can avoid the early convergence by combining the niche technology.Moreover,chaos theory is introduced to optimize the process of population initialization,and the dynamic traversal of chaos is used to select the individuals,so that the initial population is close to the target solution.On the base of studying the basic attributes of test questions and the evaluation principle of exam papers,the difficulty,knowledge and cognitive level of questions are determined as the main constraints,the mathematical model and objective function of the intelligent test paper are constructed and NCAGA is applied to the intelligent test,completing the Intelligent test paper composition of the computer system.Finally,the correctness and validity of the NCAGA method proposed in this thesis are tested by simulation experiments.The experimental results show that the NCAGA algorithm can effectively reduce the average completion time of the test paper,speed up the convergence rate of the algorithm and improve the success rate of the test paper.The algorithm has good optimization and convergence performance.On this basis,the NCAGA algorithm is applied to the undergraduate management system and student self-test platform.The results show that the quality of the test paper generated by the algorithm satisfies the actual needs of the users.