改进自适应遗传算法在关联规则中的研究

针对传统的关联规则在大数据环境下数据挖掘过程中效率低的问题,文中融入了遗传算法,并且对传统的遗传算法进行优化以便更好地适用于关联规则的挖掘。针对传统的自适应遗传算法,对其交叉算子和变异算子进行自适应优化,在此基础上加入兴趣度用来剔除一些没有实际效果的规则。实验结果表明在关联规则提取方面,结合云计算技术和所改进的算法可以优化其挖掘效率,并克服适应性和可用性等方面的缺陷,不仅加快了算法的收敛速度,也提高了解的质量。

The Study of Improved Self-adaptive Genetic Algorithm in Association Rules

Aiming at the low efficiency of traditional association rules in the data mining process under the big data environment, genetic algorithm was integrated. Meanwhile, the traditional genetic algorithm was also optimized to apply to the mining of association rules. Aiming at the traditional adaptive genetic algorithm, the crossover operator and mutation operator were optimized adaptively, and then the degree of interest was added to eliminate some rules that had no practical effect. Experimental results showed that combining cloud computing technology and the improved algorithm in the extraction of association rules could optimize its mining efficiency and overcame the shortcomings of adaptability and usability, which not only accelerated the convergence speed of the algorithm, but also improved the quality of understanding.