Article

BigHeuFIM: A Heuristic Apriori Algorithm for Extraction of Frequent Itemsets on Big Data

Anu SharmaComputer Science and Engineering Teerthanker Mahaveer University,Moradabad,IndiaAtul Kumar SrivastavaAmity University,Department of IT and Engineering,Tashkent,UzbekistanAditya Kumar TripathiComputer Science and Engineering Teerthanker Mahaveer University,Moradabad,IndiaNeeraj Kumar PandeySchool of Computing, DIT University,Dehradun,IndiaAmit Kumar MishraGraphic Era Hill University,Department of Computer Science & Engineering,Dehradun,India

2023en

ABI

Abstract

The association rule mining is the method to mine the patterns associated with the data in transactions. Setting the minimum support to a lower value increases space complexity particularly when the dataset is sizable. There are various methods that struggle to scale because of their high time complexity or memory requirements. This situation arises need for approximate results in contrast to accurate results. This work proposes a divide-and-conquer approach based heuristic strategy, BigHeuFIM, that exponentially reduce time and space complexity. The experiments demonstrates that the BigHeuFIM significantly outperformed existing algorithms in terms of speed by processing items in 80% lesser time.

Topics

Data Mining Algorithms and Applications Imbalanced Data Classification Techniques Rough Sets and Fuzzy Logic

Identifiers

DOI: 10.1109/cisct57197.2023.10351486

Citations and references

Cited by 013 references

Metrics — AkademScholar · Coming soon