Clustering in the Presence of Scatter

A new methodology is proposed for clustering datasets in the presence of scattered observations. Scattered observations are defined as unlike any other, so traditional approaches that force them into groups can lead to erroneous conclusions. Our suggested approach is a scheme which, under assumption...

Full description

Saved in:
Bibliographic Details
Published in:Biometrics 2009-06, Vol.65 (2), p.341-352
Main Authors: Maitra, Ranjan, Ramler, Ivan P.
Format: Article
Language:English
Subjects:
Algorithms
Bayes' information criterion
Biometric Methodology
Biometrics
Biometry - methods
Biostatistics
Biweight estimator
Cluster Analysis
Computer Simulation
Data Interpretation, Statistical
Datasets
Density
Epidemiologic Research Design
Estimation methods
Estimators
Exact-c-separation
Fugitives
k-clips
MCLUST
Mercury
Methylmercury
methylmercury compounds
Objective functions
Pattern Recognition, Automated
Proportional Hazards Models
Reproducibility of Results
Research methodology
Risk Assessment - methods
Sensitivity and Specificity
Statistical analysis
Statistics
Tight clustering
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A new methodology is proposed for clustering datasets in the presence of scattered observations. Scattered observations are defined as unlike any other, so traditional approaches that force them into groups can lead to erroneous conclusions. Our suggested approach is a scheme which, under assumption of homogeneous spherical clusters, iteratively builds cores around their centers and groups points within each core while identifying points outside as scatter. In the absence of scatter, the algorithm reduces to k-means. We also provide methodology to initialize the algorithm and to estimate the number of clusters in the dataset. Results in experimental situations show excellent performance, especially when clusters are elliptically symmetric. The methodology is applied to the analysis of the United States Environmental Protection Agency's Toxic Release Inventory reports on industrial releases of mercury for the year 2000.
ISSN:0006-341X
1541-0420
DOI:10.1111/j.1541-0420.2008.01064.x