Rare-events classification - An approach based on genetic algorithm and voronoi tessellation

Abdul Rauf Khan*, Henrik Schiøler, Mohamed Zaki, Murat Kulahci

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

Classification is a major constituent of the data mining tool kit. Well-known methods for classification are either built on the principle of logic or on statistical reasoning. For imbalanced and noisy cases, classification may however fail to deliver on basic data mining goals, i.e., identifying statistical dependencies in data. In this article, we propose a novel strategy for data mining based on partitioning of the feature space through Voronoi tessellation and Genetic Algorithm, where the latter is applied to solve a combinatorial optimization problem. We apply the suggested methodology to a range of classification problems of varying imbalance and noise and compare the performance of the suggested method with well-known classification methods such as (SVM, KNN, and ANN). The results obtained indicate the proposed methodology to be well suited for data mining tasks in case of highly imbalanced classes and significant noise.

Original languageEnglish
Title of host publicationTrends and Applications in Knowledge Discovery and Data Mining - PAKDD 2018 Workshops, BDASC, BDM, ML4Cyber, PAISI, DaMEMO, Revised Selected Papers
PublisherSpringer
Publication date1 Jan 2018
Pages256-266
ISBN (Print)9783030045029
DOIs
Publication statusPublished - 1 Jan 2018
Event22nd Pacific-Asia Conference on Knowledge Discovery and Data Mining - Melbourne, Australia
Duration: 3 Jun 20183 Jun 2018
Conference number: 22

Conference

Conference22nd Pacific-Asia Conference on Knowledge Discovery and Data Mining
Number22
Country/TerritoryAustralia
CityMelbourne
Period03/06/201803/06/2018
SeriesLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11154
ISSN0302-9743

Keywords

  • Classification
  • Data mining
  • Genetic algorithm
  • Imbalance
  • Noisy data
  • Voronoi tessellation

Fingerprint

Dive into the research topics of 'Rare-events classification - An approach based on genetic algorithm and voronoi tessellation'. Together they form a unique fingerprint.

Cite this