Study on EEG signal of epileptic patients based on NLSOMAP algorithm

Epilepsy is a chronic disease with sudden abnormal discharge of brain neurons, leading to transient brain dysfunction, which has a great impact on the physical and mental health of patients. In this paper, we use the NISOMAP manifold algorithm to reduce the dimension of high-dimensional data randomly selected from the public epilepsy data set, aiming at the disadvantage that the traditional supervised model is nonlinear in feature extraction. The experimental results show that, compared with other dimension reduction algorithms, NISOMAP has a better dimension reduction effect. The visualized shape is oval, with certain regularity, and it is convenient to observe the attack point. It is superior to other algorithms in different sample data, and provides great help for the diagnosis of epilepsy patients.