As stated by WHO, stress is a major problem of human beings also has a large effect on physical and mental health. This research presents one of the most basic methods for detecting stress using EEG data processing. Currents spreading through the skull of a person are created by the electrical activity of neurons in the form of voltage changes and magnetic fields, and these currents reach the scalp's surface. Voltage changes at the scalp are getting measure and this form of signals is called the EEG. These captured EEG signals got processed for obtaining useful information to identify various mental diseases. This evolved system is used to classify the characteristics of FFT and to detect whether the individual is nervous, by using the feature extraction and voting classifier. This research presents a simple and effective approach for estimating the PSD for spectrum monitoring using the Welch periodogram.. As characteristic for stress detection, the α, β, θ, & γ waves are the most appropriate characteristic. From the experiment result, we have revealed that the proposed model based on the voting classifier provides the highest 88% accuracy in comparison to a baseline model.