Analysis of State-of-The-Art Approaches for High Utility Mining: A Review

Abstract

Data mining may be outlined as an activity that extracts some new nontrivial data contained in large databases. Ancient data processing techniques have focused mostly on detecting the statistical correlations between the items that are more frequent within the transaction databases. Like frequent item set mining, these techniques are based on the rationale that item sets which appears regularly and it should be on higher priority to the user from the business perspective. In this work, we explored an emerging area referred to as utility mining that not solely considers the frequency of the item sets, however additionally considers the utility related to the item sets. In high utility item set mining the target is to identify item set that have utility values above a given threshold. Further, we tend to present a literature review on the current state of analysis on high utility mining and incorporation of genetic algorithm in data mining and also the various algorithms related to them.