LDBNN: A Local Density-based Nearest Neighbor Classifier

Joel Carbonera

2021

Abstract

K-Nearest Neighbor (KNN) is a very simple and powerful classification algorithm. In this paper, we propose a new KNN-based classifier, called local density-based nearest neighbors (LDBNN). It considers that a target instance should be classified in a class whose the k nearest neighbors constitute a dense region, where the neighbors are near to each other and also near to the target instance. The performance of the proposed algorithm was compared with the performance of 5 important KNN-based classifiers. The performance was evaluated in terms of accuracy in 16 well-known datasets. The experimental results show that the proposed algorithm achieves the highest accuracy in most of the datasets.

Paper Citation

in Harvard Style

Carbonera J. (2021). LDBNN: A Local Density-based Nearest Neighbor Classifier. In Proceedings of the 23rd International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-989-758-509-8, pages 395-401. DOI: 10.5220/0010402003950401

in Bibtex Style

@conference{iceis21,
author={Joel Carbonera},
title={LDBNN: A Local Density-based Nearest Neighbor Classifier},
booktitle={Proceedings of the 23rd International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2021},
pages={395-401},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010402003950401},
isbn={978-989-758-509-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 23rd International Conference on Enterprise Information Systems - Volume 1: ICEIS,
TI - LDBNN: A Local Density-based Nearest Neighbor Classifier
SN - 978-989-758-509-8
AU - Carbonera J.
PY - 2021
SP - 395
EP - 401
DO - 10.5220/0010402003950401