Over-sampling imbalanced datasets using the Covariance Matrix

Autor: Ireimis Leguen-deVarona, Julio Madera, Yoan Martínez-López, José Hernández-Nieto
Jazyk: angličtina
Rok vydání: 2020
Předmět:
Zdroj: EAI Endorsed Transactions on Energy Web, Vol 7, Iss 27 (2020)
Druh dokumentu: article
ISSN: 2032-944X
DOI: 10.4108/eai.13-7-2018.163982
Popis: INTRODUCTION: Nowadays, many machine learning tasks involve learning from imbalanced datasets,leading to the miss-classification of the minority class. One of the state-of-the-art approaches to ”solve” thisproblem at the data level is Synthetic Minority Over-sampling Technique (SMOTE) which in turn uses KNearest Neighbors (KNN) algorithm to select and generate new instances.OBJECTIVES: This paper presents SMOTE-Cov, a modified SMOTE that use Covariance Matrix instead ofKNN to balance datasets, with continuous attributes and binary class.METHODS: We implemented two variants SMOTE-CovI, which generates new values within the interval ofeach attribute and SMOTE-CovO, which allows some values to be outside the interval of the attributes.RESULTS: The results show that our approach has a similar performance as the state- of-the-art approaches.CONCLUSION: In this paper, a new algorithm is proposed to generate synthetic instances of the minorityclass, using the Covariance Matrix.
Databáze: Directory of Open Access Journals