论文部分内容阅读
试题组卷是考试系统的重要组成部分。通过在编码策略、适应度函数、遗传算子、控制参数等方面的研究提出一种适应于试题智能组卷的改进遗传算法。对适应度函数的适当定标和建立自适应的交叉概率和变异概率,有利于克服未成熟收敛和遗传漂移现象,同时能在维持群体多样性的情况下,防止群体进入局部最优。实验证明基本遗传算法能更有效地提高组卷的效率。
The test paper is an important part of the test system. Through the study of encoding strategy, fitness function, genetic operator and control parameters, an improved genetic algorithm is proposed to adapt to the test paper. Appropriate calibration of fitness function and establishment of self-adaptive crossover probability and mutation probability are beneficial to overcome the phenomena of immature convergence and genetic drift and prevent the population from entering the local optimum while maintaining the diversity of population. Experiments show that the basic genetic algorithm can improve the efficiency of the test paper more effectively.