基于改进遗传算法的自动组卷问题研究

Test paper auto-generation based on improved genetic algorithm

为了解决传统遗传算法在自动组卷中容易出现未成熟收敛和收敛速度慢等问题,提出了一种基于改进遗传算法的自动组卷方法。采用分段二进制编码策略,对每个子空间进行初始种群选择,保证了初始种群含有丰富的模式,从而增加搜索收敛于全局最优的可能性。并对交叉算子和变异算子进行了优化,实现了交叉和变异概率随解的变化而自适应调整。实验结果表明,改进的遗传算法能有效地解决自动组卷问题,提高了收敛速度和组卷的成功率。

In order to solve the problems prone to premature convergence and slow convergence of the traditional genetic algorithm in test paper auto-generation, a test paper auto-generation method based on an improved genetic algorithm is pro- posed. The segmented binary encoding strategy is adopted to carry out the initial population selection for each subspace to en- sure the rich patterns containing in initial population and increase the likelihood of convergence to the global optimum search. The cross operator and mutation operator were optimized. The adaptive adjustment that the crossover probability and muta- tion probability vary with the change of solution was realized. The experimental results show that the improved genetic algo- rithm can effectively solve the problem existing in the test paper auto-generation, and improve the convergence speed and the success rate of the test paper auto-generation.