Topological Decomposition and Heuristics for High Speed Clustering of Complex Networks - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2012

Topological Decomposition and Heuristics for High Speed Clustering of Complex Networks

Résumé

With the exponential growth in the size of data and networks, development of new and fast techniques to analyze and explore these networks is becoming a necessity. Moreover the emergence of scale free and small world properties in real world networks has stimulated lots of activity in the field of network analysis and data mining. Clustering remains a fundamental technique to explore and organize these networks. A challenging problem is to find a clustering algorithm that works well in terms of clustering quality and is efficient in terms of time complexity. In this paper, we propose a fast clustering algorithm which combines some heuristics with a Topological Decomposition to obtain a clustering. The algorithm which we call Topological Decomposition and Heuristics for Clustering (TDHC) is highly efficient in terms of asymptotic time complexity as compared to other existing algorithms in the literature. We also introduce a number of Heuristics to complement the clustering algorithm which increases the speed of the clustering process maintaining the high quality of clustering. We show the effectiveness of the proposed clustering method on different real world data sets and compare its results with well known clustering algorithms.
Fichier principal
Vignette du fichier
zaidi.pdf (627.53 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00679668 , version 1 (16-03-2012)

Identifiants

  • HAL Id : hal-00679668 , version 1

Citer

Faraz Zaidi, Guy Melançon. Topological Decomposition and Heuristics for High Speed Clustering of Complex Networks. Extraction et Gestion de Connaissance, 2012, Jan 2012, Bordeaux, France. pp.447-458. ⟨hal-00679668⟩
184 Consultations
146 Téléchargements

Partager

Gmail Facebook X LinkedIn More