Authors:
Luminita State
1
;
Catalina Cocianu
2
;
Ion Rosca
2
and
Panayiotis Vlamos
3
Affiliations:
1
University of Pitesti, Romania
;
2
Academy of Economic Studies, Romania
;
3
Ionian University, Greece
Keyword(s):
Feature extraction, informational skeleton, principal component analysis, unsupervised learning, cluster analysis.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Business Analytics
;
Data Engineering
;
Data Mining
;
Databases and Information Systems Integration
;
Datamining
;
Enterprise Information Systems
;
Health Information Systems
;
Sensor Networks
;
Signal Processing
;
Soft Computing
Abstract:
The aim of the research reported in the paper was twofold: to propose a new approach in cluster analysis and to investigate its performance, when it is combined with dimensionality reduction schemes. Our attempt is based on group skeletons defined by a set of orthogonal and unitary eigen vectors (principal directions) of the sample covariance matrix. Our developments impose a set of quite natural working assumptions on the true but unknown nature of the class system. The search process for the optimal clusters approximating the unknown classes towards getting homogenous groups, where the homogeneity is defined in terms of the “typicality” of components with respect to the current skeleton. Our method is described in the third section of the paper. The compression scheme was set in terms of the principal directions corresponding to the available cloud. The final section presents the results of the tests aiming the comparison between the performances of our method and the standard k-me
ans clustering technique when they are applied to the initial space as well as to compressed data.
(More)