基于SSH框架和遗传算法的多维度测评系统的设计与实现

Design and Implementation of Multi-dimensional Evaluation System Based on Framework of SSH and Genetic Algorithm

本文采用遗传算法(Genetic Algorithm,GA)实现自动组卷的系统中,并没有充分考虑试卷中知识点(测试维度)的分值分布,提出一种新的遗传算法(New Genetic Algorithm,NGA)来实现自动组卷功能,并重新定义了适应度函数,加快收敛速度。仿真实验表明,NGA算法不仅高效,且能生成一份有效试卷,使得试卷分值能在测试的维度上尽量实现均匀分布。

This paper uses the SSH framework as the system development framework to implement a multi-dimensional evaluation system. In order to achieve autogeneration of the test paper and let the score uniform distribute in every dimension,a new genetic algorithm was proposed. In the new algorithm,it redefine the fitness function to accelerate the speed of convergence. The simulation test indicates that the NGA algorithm satisfy the needs for multi-dimensional evaluation system, the speed is taster, and the quality of test paper is better.