A Global Density-based Approach for Instance Selection

Joel Carbonera

Abstract

Due to the increasing size of the datasets, instance selection techniques have been applied for reducing the computational resources involved in data mining and machine learning tasks. In this paper, we propose a global density-based approach for selecting instances. The algorithm selects only the densest instances in a given neighborhood and the instances in the boundaries among classes, while excludes potentially harmful instances. Our method was evaluated on 14 well-known datasets used in a classification task. The performance of the proposed algorithm was compared to the performances of 8 prototype selection algorithms in terms of accuracy and reduction rate. The experimental results show that, in general, the proposed algorithm provides a good trade-off between reduction rate and accuracy with reasonable time complexity.

Download


Paper Citation


in Harvard Style

Carbonera J. (2021). A Global Density-based Approach for Instance Selection. In Proceedings of the 23rd International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-989-758-509-8, pages 402-409. DOI: 10.5220/0010402104020409


in Bibtex Style

@conference{iceis21,
author={Joel Carbonera},
title={A Global Density-based Approach for Instance Selection},
booktitle={Proceedings of the 23rd International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2021},
pages={402-409},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010402104020409},
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 - A Global Density-based Approach for Instance Selection
SN - 978-989-758-509-8
AU - Carbonera J.
PY - 2021
SP - 402
EP - 409
DO - 10.5220/0010402104020409