Skip to the main content

Original scientific paper

https://doi.org/10.31341/jios.47.1.9

Data-Centric Optimization Approach for Small, Imbalanced Datasets

Vladislav Tanov ; Faculty of Economics and Business Administration, Sofia University St. Klimenr Ohridski, Sofia, Bulgaria


Full text: english pdf 620 Kb

page 167-177

downloads: 152

cite


Abstract

Data-centric is a newly explored concept, where the attention is given to data optimization methodologies and techniques to improve model performance, rather than focusing on machine learning models and hyperparameter tunning. This paper suggests an effective data optimization methodology for optimizing imbalanced small datasets that improves machine learning model performance.
This paper is focused on providing an effective solution when the number of observations is not enough to construct a machine learning model with high values of the estimated magnitudes. For example, the majority of the observations are labeled as one class (majority class), and the rest as the other, commonly considered as the class of interest (minority class). The proposed methodology does not depend on the applied classification models, rather it is based on the properties of the data resampling approach to systematically enhance and optimize the training dataset. The paper examines numerical experiments applying the data centric optimization methodology, and compares with previously obtained results by other authors.

Keywords

imbalanced dataset; classification; data centric; optimization; machine learning

Hrčak ID:

305464

URI

https://hrcak.srce.hr/305464

Publication date:

30.6.2023.

Visits: 404 *