Study of Optimization in Fragmented Item-sets for Business Intelligence

Association Rule is one of the techniques in the process data mining problems and it might be the most researched one. Some of them has developed fragment rule mining which is based on associations among large data set. Discovering item sets is the key point in fragment rule mining. Major challenge in developing fragment rule mining algorithms is the extremely large number of rules generated which makes the algorithms inefficient and makes it difficult for the end users to cope up with the generated rules. In this research, we concentrate on optimization of fragmented items sets generated from fragment rule mining. We proposed an innovative approach to find optimized association rules within inter-transaction of fragment mining. Design of this method represented in this paper which gives idea of fragmented item sets generated from fragment rule mining on which optimization is performed. This deals mainly with reducing the time and space complexity required for processing the data using fragment mining& generate strong rules using genetic algorithm. . This also reduces the width of sliding window for large data as compared to FITI because of fragmented attributes. Genetic algorithm heuristic is mainly used to generate useful solutions to optimization and search problems. In previous research many have proposed genetic approach for mining interesting association rules from large dataset. In this paper we propose knowledge based method which provides the major advancement, integrating a genetic algorithm in fragment rule mining to obtain effective rules that potentially be used for business intelligent applications.