Application Of Improved Genetic Algorithm In Auto-generating Paper

Under the background of the intelligent and prosperous development of information,the traditional teaching and examinations have gradually failed to meet the needs of modern education.The networked teaching and examination will surely be the general direction of education reform in the future.For example,in bank training,a lot of teachers edit test paper by hand or randomly.The performance of such way is low,and the quality also can not satisfy the demand.In view of the above problems and the shortcomings of the existing examination training system,this paper designs and develops a training examination system that can realize the intelligent generating test paper with the background of China Construction Bank training and examination system.The system combined genetic algorithm and swarm intelligence algorithm as an intelligent automatic test paper composition algorithm.The main work of this paper is as follow:(1)Based on the research of domestic and foreign relevant literature,and summarizes some existed problems of the current intelligent test paper composition algorithm in order to determine the research direction of this paper.(2)The related theories of genetic algorithm and swarm intelligence in intelligent training examination system are discussed,and a cluster strategy based on Web Service and Nginx and Tomcat is proposed to satisfy the training requirements of high concurrency.According to the training requirements,a general test paper adaptability evaluation model is set up.In the test paper theory,through the setting of the test paper indicators and the establishment of the mathematical model,a clear test paper strategy is established to simplify the objective function and solve the problems of uneven assignment and incomplete knowledge in the past test papers.(3)Aiming at the problem that general test paper algorithm takes more time and has lower success rate,this paper proposes a paper test papermaking algorithm based on improved genetic algorithm to solve the problem that it is hard to reach the ideal test paper.Although the global search ability of genetic algorithm is very fast,the feedback information in the system is not utilized,which leads to inefficient redundant iteration,which makes the solving efficiency lower.Therefore,improving the genetic algorithm through the fusion swarm intelligence algorithm can preserve the useful information generated in the genetic process and greatly accelerate theconvergence speed of the genetic algorithm.Finally,the improved algorithm is applied to the test paper and the simulation experiment.Both of them verify the remarkable characteristics of the algorithm with high efficiency and agility.(4)According to the bank's demand goal,the author completed the outline design and detailed design of the system,and completed the development of the system.After 6 months of normal operation,it demonstrated the system with the convenience of operation and its powerful functions,it made the bank training more intelligent,high-efficiency and economization.