Efficient Algorithms for Mining the Concise and Lossless Representation of High Utility Itemsets

Authors

  • Rakesh Patra

DOI:

https://doi.org/10.17762/msea.v70i1.2297

Abstract

Data mining is the process of extracting new, possibly useful information from vast data bases that is not straightforward. Market basket analysis, a kind of data mining used in retail research, is used to analyse client transactions. The association between the things that occur in transactions more frequently was the focus of earlier data mining techniques. They don't take an item's significance or utility into account while often mining an itemset. Utility mining is a new field that has emerged as a result of the limits of common mining goods. When mining, the profitability or utility of an object is taken into account. In a transaction, an item's utility refers to its significance or financial gain. Finding the item set with utility values over the specified threshold is the main goal of mining high utility goods. We give a literature review on several mining methods in this work. The support-confidence framework, used in traditional association rule mining, gives users an objective way to quantify the rules that are important to them. Despite several research being done, the mining operation may perform worse with regard to of processing speed as well as memory. efficiency when consumers are presented with too many high utility itemsets by existing approaches. In this research, the system proposes a unique framework for mining closed high utility itemsets that acts as a lossless yet compact representation of high utility itemsets.

Downloads

Published

2021-01-31

How to Cite

Patra, R. . (2021). Efficient Algorithms for Mining the Concise and Lossless Representation of High Utility Itemsets. Mathematical Statistician and Engineering Applications, 70(1), 173–181. https://doi.org/10.17762/msea.v70i1.2297

Issue

Section

Articles