A Generalized Association Rule Mining Framework for Pattern Discovery

Abstract: The strength of data mining is association rule mining techniques which describe the associations among items in a database and are useful to identify domain knowledge hidden in large volume of data efficiently. It is difficult to discovery association rules without providing support and confidence framework where a minimum support must be supplied to start the discovery process. In this paper, we propose a novel framework; in this associations are discovered based on logical implications. The principle of the approach considers that an association rule should only be reported when there is enough logical evidence in the data. To do this, we consider both presence and absence of items during the mining. The proposed algorithm discovers the natural threshold based on observation of data set. The different intelligence models can be used in conjunction with the proposed algorithm in determining the target item(s) to be considered during the mining process. It provides a logical underpinning to the discovery process of patterns. Currently, the illustration of the mapping of constraints to the discovery process in this paper is based on support value