A Novel Partitioning Driven Differential Evolution based Epileptic Seizure Cluster Analysis
Maninder Kaur1, Meghna Dhalaria2

1ManinderKaur, Computer Science and Engineering Department, Thapar Institute of Engineering and Technology, Patiala, India.
2Meghna Dhalaria, Computer Science and Engineering Department, Thapar Institute of Engineering and Technology, Patiala, India.

Manuscript received on 02 June 2019 | Revised Manuscript received on 10 June 2019 | Manuscript published on 30 June 2019 | PP: 1244-1249 | Volume-8 Issue-8, June 2019 | Retrieval Number: H6979068819/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Epilepsy is amongst the most widely recognized neurological ailments that reveals in redundant epileptic seizures because of an anomalous, synchronous movement of an enormous gathering of neurons.This disease comes at third level among the prevalent brain ailment that can cause detrimental effects on the day-to-day life of the victim. The early recognition of epileptic seizures is of main concern for the identification of victims with epilepsy. The present work proposes a novel partitioned based differential evolution clustering approach for detecting the epileptic seizures. The cluster process aims to arrange a collection of data samples into different clusters, in such a way that the objects belonging to a cluster are too close to one another than the objects belonging to different clusters. The partition method is primarily based on greedy heuristics approach that is used iteratively to obtain an optimal set of clusters. The proposed Partition Based Clustering using Differential Evolution (PCDE)approach is compared with Differential evolution based clustering based on cluster validity measure DB index. The empirical analysis is carried out onEpileptic Seizure medical datasets. The results revealed that PCDE approach required less computation time in comparison to DE based clustering and outperformed DE based clustering with DB Index value of 0.8626.
Keyword: Epileptic Seizure, Differential Evolution, Clustering, Partitioning.
Scope of the Article: Analysis of Algorithms and Computational Complexity.