Detection of Alcoholic using LSTM and Enhanced ANN Classification Algorithms

Around the world, alcoholism is a prevalent mental illness. Drinking too much alcohol can result in alcoholism and other problems. In extreme situations, it can result in respiratory and circulatory system paralysis and inhibition, as well as death. Furthermore, there aren't many trusted methods for detecting alcoholism. EEG signals are widely used to detect alcohol, they are data derived from observing cerebral cortex fluctuations.The majority of the machine learning approaches used in current diagnostic methods depend on human centered in order to learn. We religiously use Deep Learning as a complete approach to detect EEG signals. There aren't many research that use deep learning models to classify alcohol's EEG data. After that, a CNN and a bidirectional LSTM network are employed to extract features from the denoised data. Lastly, the classification of alcohol EEG signals is carried out. The experimental findings demonstrate that the approach suggested in this work can be used to successfully detect patients with alcoholism, outperforming the majority of existing algorithms with a diagnosis accuracy of 95.32%.