This research is related to Electroencephalography (EEG) analysis for eye movement, events feature detection, classification, hence building a fuzzy decision system for arobotic learning applications. The adopted BMI, EEG-technique, uses non-invasive technique in reading brain signals by sensors (electrodes) located at top of a skull. Works were towards EEG analysis for EYE Movements, Event Features Detection, the Recognition eye EEG brainwaves, while developing a SVM for classification of multi-related for individuals’ actions. The adopted procedures are: First the signal analysis was done, where we filtered the EEG signal and segment significant segments of the analysis and get rid of noise. Secondly, the feature extraction, where we used algorithm such as mean, average, std. deviation and more to get a specific feature which will be used in other parts of the research. The final stage is related to the classification the resulted EEG signals of the sample. In this respect, Support Vector Machine (SVM) classifier was used to identify and classifier the EEG signal into three main eye events, the (T₀, T₁, T₂ ). The purpose of this experiment is to use this technology to solve real life problems and robotics application like moving a wheelchair for disabled people, or more interactive EEG related systems.