Improving Frequent Itemset Mining Algorithms Performance Using Efficient Implementation Techniques: A Benchmark Solution

Mining frequent itemset in transactional datasets is considered to be a very challenging research oriented task in data mining due to its large applicability in real world problems. Due to the NP-complete nature of problem, the efficiency of frequent itemset mining highly depends on the efficiency of algorithm implementation. In this paper we propose a number of different implementation techniques (other than itemset mining) strategy), that can improve the running time of any frequent itemset algorithm implementation. To check the efficiency of these implementation techniques we integrate them into the original implementations of current best itemset mining implementations. We also perform our computational experiments with our modified implementations on different spare and dense benchmark datasets, which show very good results