An Efficient Clustering Based High Utility Infrequent Weighted Item Set Mining Approach

The conventional data mining techniques are mainly focused on discovering the correlation between items that are more frequent in the transaction databases. In recent years, the research area has focused on infrequent itemset mining whose frequency of occurrence is less than or equal to a maximum threshold. Most of the existing system introduced to mine infrequent item set. These methods do not consider the utility of item set. The main objective of this work is to find out the infrequent weighted item set from weighted transactional database and group them according to the utility value. An efficient high utility based clustering algorithm is used to group the high utility infrequent weighted item set by using k-mean clustering algorithm. The utility of items is determined by taking into account factors such as profit deal, temporal characteristics of items. The utility of weighted items is greater than the minimum utility threshold which is known as high Utility based Infrequent Weighted Item set. These item sets are clustered by using k-mean clustering algorithm. The proposed method achieves high performance in terms of scalability, accuracy and precision.