Fast approximate minimum spanning tree based clustering algorithm

•A fast approximate minimum tree based clustering algorithm is proposed.•A novel centroid based nearest neighbor rule is presented.•Much faster as compared to exact MST algorithms.•Experiments on both synthetic and real datasets are carried out.•Quality of clusters from the approximate MST is mainta...

Full description

Saved in:
Bibliographic Details
Published in:Neurocomputing (Amsterdam) Vol. 272; pp. 542 - 557
Main Authors: Jothi, R., Mohanty, Sraban Kumar, Ojha, Aparajita
Format: Journal Article
Language:English
Published: Elsevier B.V 10.01.2018
Subjects:
Clustering
Local neighborhood graph
Minimum spanning tree
Clustering
Minimum spanning tree
Local neighborhood graph
ISSN:0925-2312, 1872-8286
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•A fast approximate minimum tree based clustering algorithm is proposed.•A novel centroid based nearest neighbor rule is presented.•Much faster as compared to exact MST algorithms.•Experiments on both synthetic and real datasets are carried out.•Quality of clusters from the approximate MST is maintained. Minimum Spanning Tree (MST) based clustering algorithms have been employed successfully to detect clusters of heterogeneous nature. Given a dataset of n random points, most of the MST-based clustering algorithms first generate a complete graph G of the dataset and then construct MST from G. The first step of the algorithm is the major bottleneck which takes O(n2) time. This paper proposes an algorithm namely MST-based clustering on partition-based nearest neighbor graph for reducing the computational overhead. By using a centroid based nearest neighbor rule, the proposed algorithm first generates a sparse Local Neighborhood Graph (LNG) and then the approximate MST is constructed from LNG. We prove that both size and computational time to construct the graph (LNG) is O(n3/2), which is a O(n) factor improvement over the traditional algorithms. The approximate MST is constructed from LNG in O(n3/2lgn) time, which is asymptotically faster than O(n2). Experimental analysis on both synthetic and real datasets demonstrates that the computational time has been reduced significantly by maintaining the quality of clusters obtained from the approximate MST.
ISSN:0925-2312
1872-8286
DOI:10.1016/j.neucom.2017.07.038