Abstract

Itemset mining discovers interesting patterns in the dataset. The itemset may be frequent or it can be rare based on its occurrence in the database. It has been observed that most of the algorithms are designed for mining frequent itemsets. However, discovery of rare itemsets is equally important since they play a major role in making decisions in some situations. The efficiency of the algorithms depend on the way in which the data structures are designed to store and retrieve the data. Hyperlinked Rare Pattern Mining algorithm discovers all rare itemsets and is suitable for sparse dataset. In this algorithm item_id and its support count are stored in Support and Header tables. This redundancy is removed in the proposed algorithm to improve the time efficiency. An experimental analysis is conducted to discover the rare itemsets. It is observed that while there is an improvement in time efficiency, there is a tradeoff for space efficiency.

Original languageEnglish
Title of host publicationInternational Conference on Communication, Computing and Electronics Systems - Proceedings of ICCCES 2020
EditorsV. Bindhu, João Manuel Tavares, Alexandros-Apostolos A. Boulogeorgos, Chandrasekar Vuppalapati
PublisherSpringer Science and Business Media Deutschland GmbH
Pages569-577
Number of pages9
ISBN (Print)9789813349087
DOIs
Publication statusPublished - 2021
EventInternational Conference on Communication, Computing and Electronics Systems, ICCCES 2020 - Coimbatore, India
Duration: 21-10-202022-10-2020

Publication series

NameLecture Notes in Electrical Engineering
Volume733 LNEE
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

ConferenceInternational Conference on Communication, Computing and Electronics Systems, ICCCES 2020
Country/TerritoryIndia
CityCoimbatore
Period21-10-2022-10-20

All Science Journal Classification (ASJC) codes

  • Industrial and Manufacturing Engineering

Fingerprint

Dive into the research topics of 'Efficient Mining of Rare Itemsets'. Together they form a unique fingerprint.

Cite this