The goal of this research was to detect subjects with schizophrenia, a mental illness using electroencephalogram (EEG) signals. EEG is a useful technique for detecting brain activity pattern that provides significant insights regarding psychological and cognitive states. The proposed study combines many machine learning techniques, like Support Vector Machine (SVM), K-Nearest Mean (K-NN), Random Forest (RF), and XGBoost to detect schizophrenia by extracting feature from EEG signals. The study involved segmenting EEG signals, preprocessing the EEG data to identify relevant variables, training distinct algorithms for mental disorders and classifying them as healthy control and schizophrenia patient. Numerous statistical and signal handling metrics derived from EEG data are included in these feature sets. It offers crucial details regarding the characteristics of brain activity captured by EEG. When taken as a whole, these characteristics help to represent EEG data, which facilitates further assessment and categorization tasks. To boost the performance of all the above-mentioned machine learning models, hyperparameter tuning using grid search CV was performed. In conclusion, SVM provided a maximum cross-validation score of 66.75%, followed by RF (65.48%) in time domain while in frequency domain RF and XGBoost give maximum cross-validation score of 92% each. Maximum F1-score is achieved with RF and XGBoost in both the domains. This research aimed to provide neurologists with insightful information about human psychology, a practical tool for decision-making, a novel method of using EEG visuals for mental health assessment, and the capacity to take proactive steps for the preservation and enhancement of well-being.