Research And Implementation Of Intelligent Test System Based On Genetic Algorithm

With the progress of The Times,computers are being widely used by people.Especially in recent years,the rapid development of the information network,coupled with the sudden onset of the novel coronavirus epidemic,makes paperless examination become an inevitable trend.People use the computer to complete the examination process of paper grouping and marking steps,which can not only reduce the workload of teachers to issue papers and marking papers,but also reduce the use of material resources and human resources,this way can make the content of the paper more objective,the examination process more fair.Compared with the traditional examination method using paper papers,the online examination system makes use of the storage function of database,which makes the examination method has more advantages in time and space.The intelligent test paper system uses artificial intelligence,which requires the computer to automatically extract the test questions that meet people’s requirements from the question bank of the system,and then these questions form a set of test papers.In the process of designing intelligent test system,the realization of intelligent test system is a key issue that needs to be considered.The paper designs and implements an intelligent test paper composition system based on genetic algorithm because of the shortcomings of traditional test paper composition algorithm,such as low efficiency,low objectivity and poor quality of test paper.Because the traditional genetic algorithm has the problems of prematurity and slow convergence,this paper proposes to improve the genetic algorithm by using directional crossover,and improve the fitness function and constraint conditions,which can improve the ability of the algorithm to find the optimal solution.The following is the introduction of the main research work in this paper:This paper first introduces the key technologies used in the process of system development,including the characteristics of the system programming language Python,the working mechanism of Django framework,the selection of system database and architecture pattern,and gives the overall architecture of the system.Then,the paper introduces and designs the constraints of paper composition,including test paper differentiation,test paper difficulty,knowledge point coverage,etc.In the process of genetic algorithm,the coding scheme and fitness function are designed,and the genetic operator is optimized to make the paper composition more efficient and the quality of the paper better.At last,through the function and feasibility of the system needs analysis,so as to establish a complete system framework,and design the system database,database through the E-R chart and data table structure is displayed.Each function module of the system is described and designed in detail.Through the use case test of each function module of the system,it proves that the system can be used normally.