Decision support system Application for recognition of alcoholic mental State from Epileptic EEG signals

New tools to decipher complicated systems can be found in the theory of nonlinear dynamic systems. Signal processing, analysis, and classification can all benefit from its novel ideas, algorithms, and techniques. The study of the dynamics of physiological signals now frequently employs these ideas. In this study, the authors use nonlinear dynamics theory to electroencephalogram (EEG) signals in an effort to better comprehend a wide range of mental processes. In this paper, we discuss an automated method for identifying sober and drunk EEG data based on their salient properties. Several aspects are employed in the extraction process. Various ranking procedures, including the t-test, the Wilcoxon test, and the Bhattacharyya technique, were applied to the retrieved features. Classifiers like the Support Vector Machine (SVM), the Decision Tree (DT), the K-Nearest Neighbour (KNN), and the Probabilistic Neural Network (PNN) are trained using the rated features (PNN). Based on the Bhattacharyya ranking technique, the SVM classifier using a radial basis function (RBF) achieves the highest classification accuracy (96.56 percent), sensitivity (95.29 percent), and specificity (97.81 percent) for the polynomial Kernel. This system can serve as a decision support tool for doctors making alcoholism diagnoses due to its speed, accuracy, and low cost. Rehabilitation centres can benefit greatly from this method of evaluating alcoholics over time and seeing how their brain changes respond to treatment.