Litcius/Paper detail

A novel data clustering approach based on whale optimization algorithm

Tribhuvan Singh

2020Expert Systems34 citationsDOI

Abstract

Abstract Data clustering is an important technique of data mining in which the objective is to partition N data objects into K clusters that minimize the sum of intra‐cluster distances between each data object to its nearest centroid. This is an optimization problem, and various optimization algorithms have been suggested for capturing the position vectors of optimal centroids. However, in these approaches, the problem of local entrapment is very common due to weak exploration mechanism. In this paper, a novel approach based on a whale optimization algorithm (WOA) is suggested for data clustering. The performance of the suggested approach is validated using 14 benchmark datasets of the UCI machine learning repository. The experimental results and various statistical tests have justified the efficacy of the suggested approach.

Topics & Concepts

Computer scienceCluster analysisCentroidBenchmark (surveying)Data miningPartition (number theory)Optimization algorithmCluster (spacecraft)Position (finance)AlgorithmArtificial intelligencePattern recognition (psychology)Machine learningMathematical optimizationMathematicsEconomicsCombinatoricsFinanceProgramming languageGeographyGeodesyMetaheuristic Optimization Algorithms ResearchArtificial Immune Systems ApplicationsAdvanced Clustering Algorithms Research
A novel data clustering approach based on whale optimization algorithm | Litcius