Improving Classifier Accuracy for diagnosing Chronic Kidney Disease Using Support Vector Machines
C. Sathish Kumar1, P.Thangaraju2
1C.Sathish Kumar, Research Scholar, Bharathidasan University, Tiruchirappalli & Associate Professor, PG & Research Department of Computer Science, Bishop Heber College, Tiruchirappalli, (Tamil Nadu), India.
2P.Thangaraju, Associate Professor, PG & Research Department of Computer Science, Bishop Heber College, Tiruchirappalli, (Tamil Nadu), India.
Manuscript received on February 07, 2019. | Revised Manuscript received on February 14, 2019. | Manuscript published on August 30, 2019. | PP: 3697-3706 | Volume-8 Issue-6, August 2019. | Retrieval Number: F9377088619/19©BEIESP | DOI: 10.35940/ijeat.F9377.088619
Open Access | Ethics and Policies | Cite | Mendeley
© 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: Preventing Chronic Kidney Disease has become one of the most intriguing task to the healthcare society. The major objective of this paper is to deal mainly with different classification algorithms namely Naive Bayes, Multi Layer Perceptron and Support Vector Machine. The work analyzes the Chronic Kidney Disease dataset taken from the machine learning repository of UCI. Pre-processing techniques such as missing value replacement, unsupervised discretization and normalization are applied to the Chronic Kidney Disease dataset to improve accuracy. Accuracy and time are the taken as the experimental outcomes of the classification models. The final conclusion states that Support Vector Machine implements much superior than all the other classification methods.
Keywords: Classification, Preprocessing, Naïve Bayes, Multilayer Perceptron, Support Vector Machines, CKD.