Creating of Multiple Classifier Systems by Fuzzy Decision Making in Human-Computer Interface Systems

In this paper we introduce a new scheme for making decision in multiple classifier systems through a fuzzy interface system. The rules of the fuzzy interface are provided by majority vote combiner. The base classifiers we have used in the multiple classifier system is Support Vector Machines (SVM) classifiers and the difference between these classifiers is in their feature-sets and the parameters of SVM classifiers. The combination of multiple classifiers, with different features sets is performed by majority vote combination method. In order to demonstrate the performance of this multiple classifier system, we implement this system for classifying EEG signals. The task is to classify EEG signals in order to translate a subject´s intentions into a technical control signal to control the peripheral environment. We compare the individual classifiers with their ensemble counterparts and discuss the results.