Classification of EEG signals is one of the biggest problems in Brain Computer Interface (BCI) systems. This paper presents a BCI system based on using the EEG signals associated with five mental tasks (baseline, math, mental letter composing, geometric figure rotation and visual counting). EEG data for these five cognitive tasks from one subject were taken from the Colorado University database. Wavelet Transform (WT), Fast Fourier Transform (FFT) and Principal Component Analysis (PCA) were used for features extraction. Artificial Neural Network (ANN) trained by a standard back propagation algorithm and Support Vector Machines (SVMs) were used for classifying different combinations mental tasks. Experimental results show the classification accuracies achieved with the three used feature extraction techniques and the two classification techniques.