An alternative SMOTE oversampling strategy for high-dimensional datasets

Sebastián Maldonado*, Julio López, Carla Vairetti

*Autor correspondiente de este trabajo

Resultado de la investigación: Contribución a una revistaArtículorevisión exhaustiva

49 Citas (Scopus)

Resumen

In this work, the Synthetic Minority Over-sampling Technique (SMOTE) approach is adapted for high-dimensional binary settings. A novel distance metric is proposed for the computation of the neighborhood for each minority sample, which takes into account only a subset of the available attributes that are relevant for the task. Three variants for the distance metric are explored: Euclidean, Manhattan, and Chebyshev distances, and four different ranking strategies: Fisher Score, Mutual Information, Eigenvector Centrality, and Correlation Score. Our proposal was compared with various oversampling techniques on low- and high-dimensional datasets with the presence of class-imbalance, including a case study on Natural Language Processing (NLP). The proposed oversampling strategy showed superior results on average when compared with SMOTE and other variants, demonstrating the importance of selecting the right attributes when defining the neighborhood in SMOTE-based oversampling methods. © 2018 Elsevier B.V.
Idioma originalInglés
Páginas (desde-hasta)380-389
Número de páginas10
PublicaciónApplied Soft Computing Journal
Volumen76
DOI
EstadoPublicada - mar 2019

Nota bibliográfica

Publisher Copyright:
© 2018 Elsevier B.V.

Huella

Profundice en los temas de investigación de 'An alternative SMOTE oversampling strategy for high-dimensional datasets'. En conjunto forman una huella única.

Citar esto