California State University San Bernardino
Department of Computer Science and Engineering
Masters Thesis Defense

Date Wednesday, May 20, 2009 Time 3:30pm – 4:30pm Location JB-360

Title An Extendable General Platform for Cluster Analysis and Validation

Candidate Brandon Edwards

Advisor Dr. Haiyan Qiao
Committee Members Dr. Keith Schubert Dr. Ernesto Gomez

Abstract
Cluster analysis plays an important role in data analysis and knowledge 
discovery.  It is used in a large range of fields, including market 
research, city-planning, earthquake studies, and scientific research areas,
such as bioinformatics.  As the complexity and amount of data increases, 
improvement to old techniques as well as development of novel algorithms is
needed.  While cluster analysis is a useful, unsupervised technique for 
analyzing data, there are no clustering algorithms that can be uniformly 
applied to all sets of data.  Many of the algorithms need input parameters
that require the user to have some pre-existing knowledge about the data, 
such as the number of clusters the data holds.  This thesis addresses 
current problems in the area of cluster analysis, such as estimating the 
number of clusters, detecting outliers, and offering useful visualization 
of multi-dimensional data.  Solutions to these problems as well as a 
user-friendly clustering platform that aids the user in obtaining useful 
clustering results are presented.