Reduction Attributes on K-Nearest Neighbor Algorithm (KNN) using Genetic Algorithm

M. Arinal Ihsan, Muhammad Zarlis, Pahala Sirait

2018

Abstract

Diabetes mellitus (DM) is a serious health problem both in Indonesia and in the world. Data mining techniques have been done to help diagnose diabetes. Attribute selection is a process to identify and remove attributes with irrelevant or excessive values. In this study, attribute selection was performed using genetic algorithm implemented at K-Nearest Neighbor (KNN) for classification task. The genetic algorithm aims at sorting attributes by rank where the greater of an attribute the more significant the attribute for the classification task. The test was performed on Indians dataset of 768 data. From the test, we got the best combination with 1 attribute selection: 3 and 4 attribute reduction from K-Nearest Neighbor (KNN) accuracy before it was reduced 76,52%, and after reducing 76.96%. While the selection of 2 attributes is : reduce the attributes 1 and 4. The comparison of the results of K-Nearest Neighbor (KNN) accuracy before it is reduced is 76,52%, and after attribute reduction is 79,57%. These results prove that the comparison of the results obtained attribute deduction while maintaining the optimization of results before and after eliminate attributes.

Paper Citation

in Harvard Style

Ihsan M., Zarlis M. and Sirait P. (2018). Reduction Attributes on K-Nearest Neighbor Algorithm (KNN) using Genetic Algorithm. In Proceedings of the 3rd International Conference of Computer, Environment, Agriculture, Social Science, Health Science, Engineering and Technology - Volume 1: ICEST, ISBN 978-989-758-496-1, pages 371-378. DOI: 10.5220/0010043203710378

in Bibtex Style

@conference{icest18,
author={M. Arinal Ihsan and Muhammad Zarlis and Pahala Sirait},
title={Reduction Attributes on K-Nearest Neighbor Algorithm (KNN) using Genetic Algorithm},
booktitle={Proceedings of the 3rd International Conference of Computer, Environment, Agriculture, Social Science, Health Science, Engineering and Technology - Volume 1: ICEST,},
year={2018},
pages={371-378},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010043203710378},
isbn={978-989-758-496-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Conference of Computer, Environment, Agriculture, Social Science, Health Science, Engineering and Technology - Volume 1: ICEST,
TI - Reduction Attributes on K-Nearest Neighbor Algorithm (KNN) using Genetic Algorithm
SN - 978-989-758-496-1
AU - Ihsan M.
AU - Zarlis M.
AU - Sirait P.
PY - 2018
SP - 371
EP - 378
DO - 10.5220/0010043203710378