Intelligent Test Paper System Design And Implementation

With the development of computer technology,computer system is applied to the education. So intelligent test paper composition is becoming an important searching aspect. Algorithm determines the efficiency and quality of test paper construction. However, the traditional test paper composition methods have great arbitrariness and uncertainty. Except this, the complexity of time is great too. Because of the traditional methods are not intelligent, it is unable to grasp the changing requirements of the whole examination database. Taking into account of disadvantages of the existing test paper algorithm, such as low success rate, long time-consuming to generate the actual examination papers, this paper studies the issue of genetic algorithm and proposes a new test paper composition algorithm.Firstly, this paper describes the basic theories and principles of intelligence test paper composition, summarizes up the constraints during test paper the design, gives the the basic mathematical model of algorithm. Then this paper describes the genetic algorithms, included the basic idea ,the key issues and characteristics .Second, based on the theoretical analysis, this paper uses SQL database C # language and Visual Studio 2008 development environment to develop a genetic intelligent test paper system. The system has functions of user management, information management, test paper papers parameter management and intelligent test paper composition. Accordance with the theory of software engineering, based on overall system design goals, functional requirements and non-functional requirements. This paper gives the system design principles and performance requirements. After analyses system requirements, detailed design was given, including system architecture, system parameters, database design and genetic operator design. According to the user's visibility, the system is divided into the user interface layer, functional modules layer and data layers. All the ER diagram and table structure are satisfied the three paradigms of the database design requirements. In the design of Genetic operator, this paper combines the genetic algorithms and the specific of test paper composition. A real sub-encoding method was given to meet the presents test paper problems. The corresponding crossover and mutation operator were also designed. Based on the modular programming ideas, the system was consisted with user login module, system main form module, user management module, information management module, paper parameters management module and intelligent test paper module.The test indicates that the intelligent test paper based on improved genetic algorithm can satisfy the needs for actual examinations,the speed is faster and the quality of test paper is better. It had received good result and had better practicability.