Research And Implementation On Algorithm Of Paper Generating In Online Examination System

Examination has become a significant way of talent selection in today's increasingly competitive society no matter in study or work.With the rapid development of Internet technology and online education,traditional examination methods are being replaced by network exam gradually due to its own drawbacks.Since it is not controlled by space and time,greatly simplifies the workload of test paper and remarking people and more likely to store the result and information,online examination system has been adopted by more big exams.Intelligent test paper is not the core of online examination system,but the difficult in the process of system implementation.Its main function is to select a set of questions automatically from test bank that meet the relevant requirements of the test paper under certain constraints,and the result of which acts on the effect of the assessment.Therefore,the strategy of generating test paper should not only ensure the scientificity and rationality of the paper,but embrace the capability of generating test fast,conveniently and flexibly,all of which make generating test paper selection particularly important.This paper first carried on the thorough analysis of the test paper,and several important attribute indexes that affect the quality of the test paper were introduced as well as the commonly used algorithms were analyzed and compared.The genetic algorithm is widely used in test paper due to its fast,global and robustness,etc.Therefore,this paper analyzes the basic concepts of genetic algorithm and proposes HDGA,the heuristic dynamic genetic algorithm on the basis of thorough analyzing the advantages and disadvantages of specific work flow and genetic algorithm combining the features of test paper.The algorithm chooses the methods that adjust to the linear transformation fitness function,dynamic crossover and mutation probability and individual substitution.What HDGA takes could be used to optimize the population diversity problem from macroscopic and microscopic perspectives that avoid the algorithm converge to the local optimal solution.The mathematical model and the objective function are set up according to the principle of the test paper and the attributes of the test question.HDGA is applied to the problem of generating test paper,and the steps of the algorithm are designed and implemented.It could be concluded that the algorithm couldbe effectively avoid the premature convergence problem through theoretical analysis and simulation experiments.It takes online test system as platform to apply the improved genetic algorithm to intelligent test system due to the deep research and detailed analysis of HDGA algorithm.This paper has conducted a large number of experimental tests in order to further verify the effectiveness of the algorithm.And the results show that the test system based on HDGA has the merits of fast and the good quality of test paper,and which could meet the needs of the users.