基于遗传模拟退火算法的药品零售大数据关联规则挖掘

Association Rules Mining of Drug Retail Big Data Based on Genetic Simulated Annealing Algorithm

针对药品零售大数据信息,提出一种基于遗传模拟退火算法的关联规则挖掘改进算法。首先以遗传算法为主体,模拟退火算法作为其辅助,在遗传算法选择操作、交叉运算和变异运算中融入模拟退火算法,实现对算法的设计;然后运用Python语言实现了算法,并通过对药品零售大数据关联规则挖掘,发现药品零售大数据之间的关联,有效地量化了药品之间的相关程度;最后对改进算法进行有效性和可行性测试。仿真实验表明,相比遗传算法,该算法的挖掘速快,挖掘质量高,有效地提高品零售大数据关联规则挖掘的性能。

Aiming at the big data information of drug retail,an improved association rule mining algorithm based on genetic simulated annealing algorithm is proposed.Firstly,genetic algorithm is taken as the main body and simulated annealing algorithm is used as its auxiliary.Simulated annealing algorithm is integrated into the selection operation,crossover operation and mutation operation of genetic algorithm to realize the design of the algorithm.Then,the algorithm was implemented with Python language.Through mining association rules of big data of drug retail,the association between big data of drug retail was found,and the correlation degree between drugs was effectively quantified.Finally,the validity and feasibility of the improved algorithm are tested.Simulation experiment shows that compared with genetic algorithm,this algorithm can mine fast and high quality,and effectively improve the performance of big data association rules mining in retail.