Performance Comparison of Various Classifiers For Classification of Seizure

Gourav Kapoor , Prof. Pratibha Singh

As in the absence of proper technology lot of time is waste in the identification of brain signal as seizure and non-seizure. Generally a lot of tests are performed to catch the disease or to actually know whether the patient is cured or healthy. These tests results in congregate or cluster of huge number of records. Whereas many diagnostic process could result in the mesh up of the actual diagnosis process and create difficulty in obtaining the genuine result specially when there is lot of test performed. These problems could be neutralized by using classifiers for the classification of record. So there are lot of classifiers are available called as SVM (square vector machine), k-NN (k- nearest neighbours), discriminate classifier and many like these. In this study we gave a resemblance of the classifiers on the basis of their accuracy sensitivity and specificity.