Association rules mining, as the most important subject in data mining, reveals the corelations between itemsets and therefore can be widely applied to many fields such as market basket analysis, corelation analysis, classification, web-customised service, etc. Since 1993 RAgrawal, R.Srikant firstly proposed the concept of association rules, a lot of algorithms have come up in mining of association rules. While most of these algorithms are based on Apriori algorithm, will generate a huge number of candidate itemsets, need multiple scans over database, and maintain a big hash tree, so the time and space complexity is too high.This paper proposed first algorithm of frequency itemsets mining-Suppoqui. It scan database once and find frequent 1 itemsets, then, merely scaning noedsets with length of largest frequency itemsets once. Finally, finding all no redundancy itemsets.Traditional association rules mining are all based frequency itemsets and sometimes create more redundancy rules, so as to user hardly accept or reject. Therefor, this paper proposed second algorithm based association rules mining-SG. It immediately mine all no redundancy rules by avoiding find frequency itemsets.No matter what time and space, Suppoqui and SG decrease according to quantitative level with traditional algorithms based Apriori. So, they have higher efficiency and feasibility.
|