Introduction: This research focuses on neural networks based biological signal processing to solve the complex classification problems. Many types of research of classification algorithms have been published, but none has effectively focused on implementing them in brain Epileptic Seizure Electroencephalography pattern analyses and lobe classification. Objective: To develop different autoregression feature extraction algorithms for identifying accurate features in the epileptic seizure EEG signals for the neural network-based classification. Methods: In this research, the Probabilistic Neural Network (PNN) is considered for classifying the brain tissue samples by mapping the input pattern to several classifications. The dataset is retrieved from the Karunya University Epileptic Seizure Database for verifying the experiment with 10-20 electrodes. Different mental tasks are considered here to verify the proposed Probabilistic Neural Network-based Epileptic Lobe Seizure classifier. Results: The experiments are carried out with several Auto Regression features. Further, the obtained result proves that the proposed PNN model has a maximum accuracy of 96.30%. Conclusion: This research work has aimed to design a PNN classifier for detecting Seizure by incorporating AR parametric features. The proposed system bears the potential of providing an exact identification of faults and noise with various age criteria. It will help process the data in a user-friendly manner.