Litcius/Paper detail

A Brief Comparison of K-means and Agglomerative Hierarchical Clustering Algorithms on Small Datasets

Hassan I. Abdalla

2022Lecture notes in electrical engineering16 citationsDOIOpen Access PDF

Abstract

Abstract In this work, the agglomerative hierarchical clustering and K-means clustering algorithms are implemented on small datasets. Considering that the selection of the similarity measure is a vital factor in data clustering, two measures are used in this study - cosine similarity measure and Euclidean distance - along with two evaluation metrics - entropy and purity - to assess the clustering quality. The datasets used in this work are taken from UCI machine learning depository. The experimental results indicate that k-means clustering outperformed hierarchical clustering in terms of entropy and purity using cosine similarity measure. However, hierarchical clustering outperformed k-means clustering using Euclidean distance. It is noted that performance of clustering algorithm is highly dependent on the similarity measure. Moreover, as the number of clusters gets reasonably increased, the clustering algorithms’ performance gets higher.

Topics & Concepts

Hierarchical clusteringCluster analysisSingle-linkage clusteringCorrelation clusteringPattern recognition (psychology)Euclidean distanceCURE data clustering algorithmData miningConsensus clusteringComputer scienceEntropy (arrow of time)Similarity measureFuzzy clusteringArtificial intelligenceMathematicsComplete-linkage clusteringCanopy clustering algorithmSimilarity (geometry)Hierarchical clustering of networksMeasure (data warehouse)PhysicsImage (mathematics)Quantum mechanicsAdvanced Clustering Algorithms ResearchData Mining Algorithms and ApplicationsCustomer churn and segmentation