The Design And Realization Of The Intelligent Test Paper Model For College English Examination

Exams are widely used as the most commonly used test method,and the quality of test papers has a great impact on the scientificity and fairness of the exam.At present,test questions are composed mainly by teachers manually selecting test questions from the test question bank to realize test paper composition,but this manual method has problems such as large manpower consumption and the inability to guarantee the quality of test papers.With the rapid growth of test papers and student examination data,as well as the development of artificial intelligence technology,intelligent paper composition methods such as random method,simulated annealing algorithm and genetic algorithm have attracted attention.Aiming at the defect that the genetic algorithm is easy to fall into the local optimal solution,this paper improves the genetic algorithm and proposes a test paper algorithm based on the diseased enhanced genetic algorithm.First,the attributes of the college English test questions are defined,and the mathematical model of the test questions is constructed;the teacher’s target requirements for the test paper are converted into 8 constraints including the total score constraint,and the objective function is defined as the sum of the weighted errors of the test paper on each constraint.Then,the genetic algorithm is optimized,and the chromosome is divided into 4 sections according to the different types of questions by using the segmented real number coding method,and the test question number is used as the identification of each gene.Adaptively changing crossover probability is adopted in the crossover operator,the higher the crossover probability is when the diversity of the population is better,the lower the crossover probability is when the fitness of the paired parties is higher.In view of the fact that different test questions may have the same attributes,the improved Hamming distance is used to calculate the distance between individuals in the population.If the attributes of two genes are exactly the same except for the test question number and exposure,the two genes are regarded as the same gene.A disease operator is designed to replace the mutation operator,which includes three modules: natural illness,disease infection,and mutation recovery.The niche technology based on the elimination strategy is introduced,and when the distance between individuals is too close in the population,the individuals with low fitness are punished.In order to verify the performance of the proposed paper-making algorithm based on the diseased enhanced genetic algorithm,the college English test questions from 2014 to 2021 were used as the data set.According to the four evaluation indicators of the success rate of test papers,the total number of iterations,the average distance of the population,and the average fitness of the population,comparative experiments were carried out on the test paper algorithms based on random sampling method,the test paper algorithm based on genetic algorithm,the test paper algorithm based on improved genetic algorithm,and the test paper algorithm based on diseased enhanced genetic algorithm respectively.The experimental results show that the test paper algorithm based on diseased enhanced genetic algorithm has stronger global solution ability and higher efficiency,which can effectively improve the success rate of test paper and reduce the number of iterations.Finally,this paper implements the intelligent paper composition model for college English examinations.When teachers use custom parameters to assemble papers according to teaching needs in the paper composition interface,or directly call templates to make papers,this model will quickly and efficiently generate a test paper and provide two options of reorganizing the test paper with the same parameters and saving the test paper locally.