Epilepsy is a chronic neurological disorder of the brain, approximately 1% of the world population suffers from epilepsy. Epilepsy is characterized by recurrent seizures that cause rapid but revertible changes in the brain functions. Temporary electrical interference of the brain roots epileptic seizures. The occurrence of an epileptic seizure appears unpredictable. Various methods have been proposed for dimensionality reduction and feature extraction, such as Principal Component Analysis (PCA), Independent Component Analysis (ICA), and Singular Value Decomposition (SVD). The advantages of the regularization dimension are that it is more precise than other approximation methods and it is easy to derive an estimator in the presence of noise due to the fully analytical definition. This paper gives an overall review of the dimensionality reduction techniques suitable for the application of electroencephalography signals from epileptic patients.

In recent years, a number of researchers have concentrated on medical data analytics because machine intelligence in medical diagnosis is a new trend for enormous medical applications. Generally, medical datasets are massive in size, so traditional classifiers suffered from overfitting and under-fitting problem of training set. In this paper, Gradient Descent Logistic Regression (GDLR) classification method is proposed for medical data classification. The Pearson Correlation Coefficient (PCC) is used to calculate the correlation between the features. After that, Random Forest (RF) algorithm ranks the features and selects the most relevant features to improve performance of the medical data classification. The regression technique processes the features effective and analyse the feature importance based on the weight values. The Random Forest (RF) assigns the features importance in the tree structure. The random forest is used to select the features and features are applied for the GDLR to classify effectively. The GDLR method further analysis the features for effectively analysis the feature importance based on the weight values and more relevant features are identified than the RF. The experimental analysis demonstrated that the performance of GDLR algorithm achieved better than traditional methods Neural Network for Threshold Selection (NNTS) and Mean Selection (MS). The accuracy of the proposed GDLR method achieved as 97.5% in the Hepatitis dataset, while existing mean selection method has the accuracy of 82.58%.