摘要
组卷模块能否高效率地生成符合系统要求的试卷是评估考试系统的重要因素。遗传算法(GA算法)近年来在解决多重约束问题的求解中应用非常广泛。把遗传算法应用于自动组卷,在标准遗传算法的基础上进行了改进,采用分段编码,基于误差的适应度函数和动态交叉及变异概率。采用遗传算法的智能组卷系统既能很好的满足组卷的约束,且效率远高于传统的回溯算法。
It is one of the major factors to evaluate a examination system that whether its testpaper generation module can produce the testpapers as the system requirements effectively. Genetic algorithms has been widely used in solving the multi-constrains problems recently. Apply genetic algorithms in auto-generating testpaper and make imporvement to the standard genetic algorithms by using segmental encoding,adaptation funcion based on error and dynamic probability of cross ＆ variation operations. The intelligent testpaper generation system incorporating genetic algorithms not only can satisfy the required constrains but also manifest far more better efficiency over testpaper generation systems which adopt traditional trace algorithms.
出处
《价值工程》
2010年第14期159-161,共3页
Value Engineering
关键词
遗传算法
考试系统
自动组卷
genetic algorithms
examination system
auto-generating testpaper
