Binary Classification of Epilepsy Using Classical Machine Learning Algorithms and Ensemble Learning Techniques

Epilepsy is a neurological disorder characterized by recurrent seizures. Seizures are caused by abnormal electrical activity in the brain. Symptoms of epilepsy can vary widely, depending on the type of seizure and the part of the brain affected. Some people with epilepsy may experience only occasional seizures, unconsciousness, and temporary confusion while others may have frequent seizures that are difficult to control. Early detection of Epilepsy is possible through several form of input data, ranging from Video, Audio, Text, several physiological data (such as Heartbeat & Blood pressure) and ECG. In this paper we have used ECG recoding data and several Machine Learning algorithms to detect Epilepsy. In our Experiment, we have used ‘UCI Epileptic Seizure Recognition Dataset’ and to counter the Problem of Imbalance data set we have used SMOTE analysis. We have also done various Hyper parameter tuning processes to improve the overall result of the performance of the model. We have found that the Gradient Tree Boosting Algorithm Has Outperformed K- Nearest Neighbor (K-NN), Naive Bayes, Logistic Regression, Decision Tree (D.T.), Random Tree classifiers with 97.41% Accuracy, ROC = 0.974, and RMSE = 0.16.