CAJ | 학술논문

针对传统遗传算法收敛速度慢、稳定性能差的缺陷，根据云计算思想提出一种遗传算法的考试系统组卷算法。该算法利用正态云模型的随机性和倾向性，动态调整遗传算法的个体选择适应度值和交叉概率和变异概率，以加快算法向最优解的逼近速度，可以在试题库中按照试题类型、试题数量、曝光度等约束条件进行快速搜索，系统通过选择、交叉和变异等操作，从试题库中自动地查找和组织出一些不同类型、不同难度、不同章节范围的试题来组成一套最佳的试卷，实现了快速自动组卷功能。
침대전통유전산법수렴속도만、은정성능차적결함，근거운계산사상제출일충유전산법적고시계통조권산법。해산법이용정태운모형적수궤성화경향성，동태조정유전산법적개체선택괄응도치화교차개솔화변이개솔，이가쾌산법향최우해적핍근속도，가이재시제고중안조시제류형、시제수량、폭광도등약속조건진행쾌속수색，계통통과선택、교차화변이등조작，종시제고중자동지사조화조직출일사불동류형、불동난도、불동장절범위적시제래조성일투최가적시권，실현료쾌속자동조권공능。
Aiming at the defects that the traditional genetic algorithm convergence speed ,stability is poor ,based on the concept of cloud computing algorithm ,examination system for genetic algorithm is proposed .The algorithm uses the ran-domness and the tendency of the normal cloud model ,individual choice for the adjustment of genetic algorithm with dynamic fitness value and the crossover probability and mutation probability ,to speed up the algorithm convergence to the optimal so-lution ,can be in the test database according to the test question types ,quantity ,visibility constraints for rapid search ,sys-tem through selection ,crossover and mutation operation ,automatically search from the test database and organize a number of different types ,different sections ,different difficulty range the database to form a set of best papers ,to realize fast auto-matic test paper .