Title Author-s Contact lnfo Department Major citation Year of Publication Publisher Sponsor Type of Publication ISSN Analytical Evaluation of hybrid classification methods for EEG based Brain Computer Interface Ijaz Ali Shoukat, M. Iftikhar, mohsinunsw@gmail.com Computer science Accepted in WORKSHOP ON UNESCO-HP “BRAIN GAIN INITIATIVE” In Conjunction with 2nd Kuwait Conference on E-Services & E-Systems, April 5-7, 2011, Kuwait University, Kuwait. Proceedings of ACM Inc. 2011 2011 2nd Kuwait Conference on E-Services & E-Systems Conference URI/DOI Full Text (Yes,No) Key words Abstract No Joint Hybrid Classification Methods, Learning Classifiers, Brain Computer Interface, EEG Significance: Brain Computer Interfacing research has made possible that a machine can be controlled by human thinking with such a passion we just think and tasks are done by the machine without any physical effort. Classification methods are the backbone of any kind of BCI Frame work especially for Electroencephalographic signal’s based systems. Motive: Electroencephalographic signals contain variety of problems like noise, multi dimensionality, outlier, non- stationary, varying size of training sets, time variations and artifacts that cause diverse training and classification to make less accurate decision. Hybrid classifier is better than single classifier for getting effective classification accuracy. Many hybrid classifiers persist but it is a question mark which combination of classifiers is better. This paper provides analytical comparison under theoretical set of rules to conclude which hybrid method can provide wrathful learning and classification services for EEG based Brain Computer Interfacing systems. Approaches: We implemented quantitative grading criteria which follows analytical comparison under theoretical set of rules to judge the hypothetical capability of different joint assembly of classifiers. Findings: Joint Hybrid Method (HMM + SVM + LDA + CSP) is superior to overcome maximum capacity of problems related to learning and classification of EEG signals for EEG based Brain Compute Interface.