Relative density-based clustering algorithm for identifying diverse density clusters effectively

Clustering is an important part of data mining. The existing clustering algorithm failed in the data set with uneven density distribution. In this paper, we propose a novel clustering algorithm relative density-based clustering algorithm for identifying diverse density clusters effectively called ID...

Full description

Saved in:
Bibliographic Details
Published in:Neural computing & applications Vol. 33; no. 16; pp. 10141 - 10157
Main Authors: Wang, Yuying, Yang, Youlong
Format: Journal Article
Language:English
Published: London Springer London 01.08.2021
Springer Nature B.V
Subjects:
Algorithms
Artificial Intelligence
Clustering
Computational Biology/Bioinformatics
Computational Science and Engineering
Computer Science
Data mining
Data Mining and Knowledge Discovery
Data points
Datasets
Density
Density distribution
Image Processing and Computer Vision
Original Article
Outliers (statistics)
Probability and Statistics in Computer Science
Allocation strategy
nearest neighbors
Relative density
Density-based clustering
ISSN:0941-0643, 1433-3058
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Clustering is an important part of data mining. The existing clustering algorithm failed in the data set with uneven density distribution. In this paper, we propose a novel clustering algorithm relative density-based clustering algorithm for identifying diverse density clusters effectively called IDDC. It can effectively identify clusters in data sets with different densities and can also handle outliers. We first compute relative density for each data point. Then, the density peak points are screened and the initial clusters are obtained according to these peak points. The strategy for assigning the remaining points is to find unallocated points from the perspective of the cluster, which can effectively identify different density. In experiments, we compare the proposed algorithm IDDC with some existing algorithms on synthetic and real-world data sets. The results show that IDDC performs better than those existing algorithms, especially clustering on data set with uneven density distribution.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0941-0643
1433-3058
DOI:10.1007/s00521-021-05777-2