Principal component analysis implementation for brainwave signal reduction based on cognitive activity

Human has the ability to think that comes from the brain. Electrical signals generated by brain and represented in wave form. To record and measure the activity of brainwaves in the form of electrical potential required electroencephalogram (EEG). In this study a cognitive task is applied to trigger a specific human brain response arising from the cognitive aspect. Stimulation is given by using nine types of cognitive tasks including breath, color, face, finger, math, object, password thinking, singing, and sports. Principal component analysis (PCA) is implemented as a first step to reduce data and to get the main component of feature extraction results obtained from EEG acquisition. The results show that PCA succeeded reducing 108 existing datasets to 2 prominent factors with a cumulative rate of 65.7%. Factor 1 (F1) includes mean, standard deviation, and entropy, while factor 2 (F2) includes skewness and kurtosis.