基于差分进化算法的智能组卷研究

Intelligent test-sheet composition research based on differential evolution algorithm

符合实际组卷需求的组卷问题模型和高效优质的组卷算法是实现计算机自动组卷的关键。针对当前计算机自动组卷模型在重要组卷指标上存在误差的现状,提出了一种能够满足用户精确组卷需求的改进组卷问题模型。将新出现的差分进化算法应用于所提出的模型,给出了一种新型智能组卷算法。利用不同规模的真实题库,进行了算法的模拟实验。实验结果表明,与基本遗传算法相比,该算法在组卷成功率和组卷质量方面具有更好的性能。

The problem model that satisfied with the practical requirements and efficient test-sheet composition algorithm with high performance play key roles for computer-aided test system. Current test-sheet composition model can＇t meet the accurate demands of users in some important test-sheet characters. In order to overcome the drawbacks of the current models, an improved test-sheet composition model is proposed. Based on differential evolution algorithm, a novel intelligent test-sheet composition algorithm is designed. In simulation experiments, the proposed algorithm is applied to a range of real item-banks. Superiority of the proposed algorithm in success rate and test-sheet quality is demonstrated by comparing it with the simple genetic algorithm.