Correlation Dimension Based Performance Analysis of Alcoholic EEG Data with PCA and PSO Classifiers
Harikumar Rajaguru1, Vigneshkumar Arunachalam2
1Harikumar Rajaguru, Professor, Department of Electronics and Communication Engineering, Bannari Amman Institute of Technology, Sathyamangalam (Tamil Nadu), India.
2Vigneshkumar Arunachalam, Assistant Professor, Department of Computer Science and Engineering, Bannari Amman Institute of Technology, Sathyamangalam (Tamil Nadu), India.
Manuscript received on 17 December 2018 | Revised Manuscript received on 29 December 2018 | Manuscript Published on 24 January 2019 | PP: 403-406 | Volume-7 Issue-4S2 December 2018 | Retrieval Number: ES2093017518/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Excessive drinking Alcohol is a serious problem in the world as it causes a lot of health issues. It causes heavy damage in the human brain and problems like lack of memory and concentration accompanied by impaired decision making takes place. The electrical activity of human brain is identified by, Electroencephalography (EEG) signals. An EEG signal depicts multiple patterns for various neurological disorders. Hence, it is widely preferred one in clinical diagnosis. A chronic alcoholic patient’s EEG and the Correlation Dimension (CD) features are analyzed. The obtained CD features are classified with the Principle Component Analysis (PCA) and Particle Swarm Optimization (PSO) Classifiers. The bench mark parameters such as Good Value(GV), AUC, Specificity and Sensitivity are compared in both optimization. The PSO Classifier out performed PCA Classifier with higher AUC of 97.92% when compared with PCA’s AUC of 96.85%.
Keywords: EEG, Correlation Dimension, Particle Swarm Optimization (PSO), Principle Component Analysis (PCA).
Scope of the Article: Data Warehousing