Combining Apriori and FP Growth algorithms with Simulated Annealing for Optimized Association Rule Mining

Abstract: Association rule mining is the process of finding frequent patterns and associations between set of objects from information repositories. Finding optimized techniques for generating association rules from large repositories has become a major area of study. Apriori algorithm is a simple algorithm which is used for mining frequent item sets. The FP-growth algorithm on the other hand works as a solution to the problem for long frequent patterns to searching for shorter ones recursively. The output of FP-Growth is a FP Tree in the end. The current work focuses on using Simulated Annealing technique on both algorithms for optimized Association Rule Mining. The results have been also discussed in the end and analysis has also been generated.

Keywords: Apriori, FP-Growth, Simulated Annealing, Minimum Support, Minimum Confidence.