A strategy for mining utility based temporal association rules

Due to the widespread computerization and affordable storage facilities, there exists enormous amount of information in databases belonging to different enterprises. The ultimate intent of this massive data collection is the utilization of this to achieve competitive benefits, by determining formerly unidentified patterns in data that can direct the process of decision making. Data mining, the core step of Knowledge Discovery in Databases (KDD) is the process of applying computational techniques that, under acceptable computational efficiency limitations, produce a particular enumeration of patterns. Data mining tasks can be classified in to two categories. Descriptive mining and Predictive mining. Association Rule mining (ARM), Clustering and sequential pattern mining are some of the descriptive mining. The main advantages of Association rules are simplicity, intuitiveness and freedom from model-based assumptions. In this paper, an extensive survey of Association rule in regard to temporal databases and utilities are done. The proposed algorithm is able to mine temporal association rules based on utilities by adapting the support with relevant to the time periods and utility. An approach of mining UTARM is designed and the efficiency of the method is discussed.