基于改进遗传算法的自动组卷系统的研究

A Study of the Automatic Test System Based on Improved Genetic Algorithm

针对基本的遗传算法在自动组卷系统中容易陷入局部最优解、迭代后期容易早熟收敛等缺点,提出了改进的初始种群选择方法、自适应的交叉概率和变异概率的改进遗传算法。并且通过对组卷数学模型的改进,使得系统对多门课程具有通用性。实验结果表明,改进遗传算法改善了算法的全局搜索能力,更好地克服了迭代后期的早熟现象,因而在组卷效果及效率上优于基本遗传算法。

In view of the shortages, which the basic genetic algorithm in automatic test paper generation system is easy to fall into local optimal solution and iterative later easily prcmature convergence, the method of improving the initial population selection, the adaptive crossover probability and mutation probability of the improved genetic algorithm is proposed. The experimental results show that the improved genetic algorithm in the test effect and efficiency is better than the basic genetic algorithm.