基于改进遗传算法的自动组卷问题的研究

Research of automatic test paper composition based on improved Genetic Algorithm

自动组卷优化问题的研究,由于题库组卷的随机性,难度很大。针对传统遗传算法在自动组卷中存在的未成熟收敛和收敛速度慢等问题,为了快速可信地组卷,提出了一种改进的遗传算法。该算法采用模拟小生境法选择算子进行种群选取,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的非线性自适应调整。进行仿真实验,结果证明,改进的遗传算法在组卷的有效性、稳定性和收敛速度等方面有显著的提高,更能有效解决自动组卷问题,具有较好的使用性能和实用性,能够极大的满足用户组卷的需求。

Research of automatic generation of optimization problems, due to the randomness of the exam test paper, it is very difficult. In order to quickly test paper credibly ,proposes an improved genetic algorithm aiming at dealing with the problems of premature convergence and low converging speed. The algorithm adopts a new method that simulated niche is introduced in the selection operator for the population selection and makes the crossover probability and mutationprobability adjust adaptively and nonlinearl.Experiment proves that the effectiveness, stability and converging speed have been promoted in the improved genetic algorithm, which could settle problem of auto-composing test paper more effectively and has good performance and practicability, and satisfy the needs of auto generating examination papers.