Classification of Alcoholic and Non-Alcoholic EEG Signals Based on Sliding-SSA and Independent Component Analysis

Alcoholism is a widely affected disorder that leads to critical brain deficiencies such as emotional and behavioural impairments. One of the prominent sources detect alcoholism by analysing Electroencephalogram (EEG) signals. Previously, most works have focused on detecting using various machine deep learning algorithms. This paper has used novel algorithm named Sliding Singular Spectrum Analysis (S-SSA) decompose de-noise EEG We considered independent component analysis (ICA) select alcoholic non-alcoholic components from preprocessed data. Later, these were train test models like SVM, KNN, ANN, GBoost, AdaBoost XGBoost classify The sliding SSA-ICA helps in reducing computational time complexity models. To validate performance ICA algorithm, we compared accuracy with its counterpart, principal (PCA) linear discriminant (LDA). proposed tested publicly available UCI dataset. verify models, calculated metrics accuracy, precision, recall F1 score. Our work reported highest 98.97% classifier. validation method done comparing classification latest state-of-the-art works.