Reduction of Association Rules for Big Data Sets in Socially-Aware Computing
Woo Sik Seol, Hwi Woon Jeong, Byungjun Lee, Hee Yong Youn · 2013
Reduction of the number of association rules in data mining is a very important issue in the field of socially-aware computing in which big data need to be manipulated. The existing schemes based on the frequency of occurrences are not effective for relatively large size dataset. In this paper we propose the tabular-algorithm that assigns a weight to each rule for the removal of unimportant rules and employs the Quine-Mccluskey method for rule reduction. Computer simulation reveals that the proposed scheme significantly improves support, credibility, rule reduction rate, and processing time compared to the representative existing schemes such as Apriori and FP-growth algorithm.