Affiliation:
1. Govt. K. P.G. College Jagdalpur Bastar, Jagdalpur, India
2. Dr. C. V. Raman University Kota Bialspur (C.G.), Bilaspur, India
Abstract
The purpose of this article is to weigh up the foremost imperative features of Chronic Kidney Disease (CKD). This study is based mostly on three cluster techniques like; K means, Fuzzy c-means and hierarchical clustering. The authors used evolutionary techniques like genetic algorithms (GA) to extend the performance of the clustering model. The performance of these three clusters: live parameter purity, entropy, and Adjusted Rand Index (ARI) have been contemplated. The best purity is obtained by the K-means clustering technique, 96.50%; whereas, Fuzzy C-means clustering received 93.50% and hierarchical clustering was the lowest at 92. 25%. After using evolutionary technique Genetic Algorithm as Feature selection technique, the best purity is obtained by hierarchical clustering, 97.50%, compared to K –means clustering, 96.75%, and Fuzzy C-means clustering at 94.00%.
Reference27 articles.
1. A novel approach for outlier detection and clustering improvement
2. FCM: The fuzzy c-means clustering algorithm
3. A Classification of CKD Cases Using MultiVariate K-Means Clustering.;A.Dubey;International Journal of Scientific and Research Publications,2015
4. A Fuzzy Relative of the ISODATA Process and Its Use in Detecting Compact Well-Separated Clusters
5. An improved of spam E-mail classification mechanism using K-means clustering.;N. O. F.Elssied;Journal of Theoretical and Applied Information Technology,2014