Improved pseudo-association rules technique

Various data mining methods are being considered to find the frequent itemsets from large data repository. In this paper an effort has been made to not only examine the past and present transaction database to discover customer purchasing behavior but also to find the strength of association relation between the data items. Association rule is used to find the rules from the past transaction data, but as the time goes on and the environment or customer choice changes, association rules of past transaction data items might become obsolete. Hence rules should be checked regularly and categories according to the strength of measures. In this paper, a new algorithms `Improved Pseudo-association Rules' (abbreviated IPAR)is proposed, which finds the strength of the relations between the data items. The advantage of this proposed algorithm is that it finds the strength of the association rules and also categories it into Unreasonable, Reasonable and Strong subset in less time. This categorization can help to understand which association rule has more happening or which has less happening. The experimental result shows that the association rule generated by present algorithm has reasonably lesser execution time and strength of the relationships. It also shows the trend of association rules of each category.