Online Examination System Of Automatic Paper Composition Based On Hybrid Algorithm

In 2018,the "Education Informatization 2.0 Action Plan" issued by the Ministry of Education clearly proposed the development goal of smart education for the first time.At the same time,in the context of COVID-21,education and teaching have been transformed to the network,among which online examination mode has gradually become an important way for many universities and enterprises to evaluate talents.Compared with the traditional examination mode,which requires the examiners to choose the topic,mark the paper and count the results,the online examination system can greatly save the material resources and manpower,and the network transmission and storage of the examination paper can reduce the leakage of the examination information and the unfair situation of the examination.The core of the online examination system is the automatic paper composition algorithm,but the traditional automatic paper composition algorithm has some problems,such as low efficiency,high randomness,difficult to control the paper and so on.In view of the above problems,this paper analyzes the related theories of automatic paper composition,and designs a comprehensive and superior performance based on hybrid algorithm automatic paper composition online examination system.The main work of this paper is as follows:(1)Multi-keyword retrieval based on test site knowledge graph:For the group in the process of exam does not match the search time consuming problem again,put forward to establish the test method of knowledge map,test questions from the question bank entity according to certain way,promoted in the process of similar subject retrieval efficiency,at the same time,through the knowledge fusion technology to establish the subgraph diagram,similar topic,reduce the retrieval time(2)Automatic paper making based on improved genetic algorithm:Aiming at the problems of low efficiency and high randomness of traditional paper making algorithm,an automatic paper making based on improved genetic algorithm with overall category adaptation is designed.The constraint conditions of test paper content are mapped to the expression space of genetic algorithm.In order to avoid premature convergence of genetic algorithm and fall into local optimum,the paper quality is improved by piecewise coding of questions according to database table relations.In the iterative process of the algorithm,dynamic adaptive crossover and mutation operators and elite reservation strategies of subgroups are used to avoid premature convergence of genetic algorithm and to improve the quality of test papers.(3)Analyze and design the automatic test paper online examination system:determine the core functions through the system demand analysis.According to the user identity,it can be divided into three categories:administrator,teacher and student.According to the functional requirements,it can be designed into four functional modules,including login management,automatic paper composition,test management and online examination module.Then,the flow chart of each module is analyzed.After determining the overall requirements and design of the system,the analysis and design of the system database are completed.