Authors:
André Lourenço
1
and
Ana Fred
2
Affiliations:
1
Instituto de Telecomunicacoes, Instituto Superior de Engenharia de Lisboa, Portugal
;
2
Instituto de Telecomunicacoes, Instituto Superior Tecnico, Portugal
Keyword(s):
Visualization, Unsupervised Learning, Clustering, Evidence Accumulation Clustering, Co-association Matrix, ECG Analysis, Behavior Identification.
Related
Ontology
Subjects/Areas/Topics:
Applications
;
Artificial Intelligence
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Biometrics
;
Biometrics and Pattern Recognition
;
Data Manipulation
;
Health Engineering and Technology Applications
;
Human-Computer Interaction
;
Methodologies and Methods
;
Multimedia
;
Multimedia Signal Processing
;
Neurocomputing
;
Neurotechnology, Electronics and Informatics
;
Pattern Recognition
;
Physiological Computing Systems
;
Sensor Networks
;
Soft Computing
;
Telecommunications
Abstract:
The representation of data in some visual form is one of the first steps in a data-mining process in order to gain some insight about its structure. We propose to explore well known visualization and unsupervised learning techniques, namely clustering, to improve the understanding about the data and to enhance possible relations or intrinsic similarity between patterns. Specifically, Clustering Ensemble Methods are exploited separately and combined to provide a clearer visualization of data organization. The presented methodology is used to improve the understanding of ECG signal acquired during Human Computer Interaction (HCI).