基于社会计算的大数据集关联规则的研究

摘要

针对出现大规模的数据集无效的问题,提出了一种新的挖掘方法WTabular算法.该算法为每条规则分配一个权重,移除不重要的规则,结合奎因-麦克拉斯基算法来对规则进行简化.实验表明,与传统的代表性算法,如APRIORI算法和频繁模式(FP)增长算法相比,提出的WTabular方法有效地改善了支持度、可靠性,规则简化率以及处理时间.%The simplification of the association rules of data mining is a very important topic in the field of social computing,and the problem of the existing scheme of frequency is not valid for the relatively large data sets.A new method for mining WTabular algorithm is proposed.The algorithm assigns a weight for each rule,removes less important rules and combines with the Quine McCluskey algorithm of rules to simplify rules.Experiments show that compared with the traditional representation al gorithms,such as APRIORI algorithm and frequent pattern (FP) growth algorithm,this method can effectively improve the support degree,reliability,rule reduction rate and processing time.