摘要
对数据挖掘的关联规则数目进行简化是社会化计算领域的一个非常重要的话题,针对出现频率的现有方案对相对大规模的数据集无效的问题,提出一种新的挖掘方法 WTabular算法,该算法给每条规则分配一个权重,移除不重要的规则并结合奎因-麦克拉斯基算法来对规则进行简化。实验表明相比传统的代表性算法,如APRIORI算法和频繁模式(FP)增长算法,本文方法有效地提高了支持度、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. In this paper, 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. Experiments show that compared with the traditional representation algorithms, 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.
出处
《佳木斯大学学报(自然科学版)》
CAS
2017年第3期396-400,共5页
Journal of Jiamusi University:Natural Science Edition
基金
四川省教育厅重点项目(15ZA0339)
阿坝师专校级规划项目(ASB12-24)