A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Performance of Sampling Methods on Imbalanced Data: Comparative Analysis
Imbalanced data affects a range of machine learning applications including bioengineering. Sampling is the most common approach to deal with imbalanced data. In this study, the effectiveness of various popular sampling algorithms that are used to balance imbalanced data were compared. Concretely, five popular sampling algorithms were considered - synthetic minority oversampling technique, random oversampling, adaptive synthetic sampling, Gamma, and random undersampling - in dealing with imbalanced data. The performance of three machine learning algorithms was evaluated - random forest, support vector machines, and deep neural network - on ten imbalanced datasets that are artificially balanced. Through multiple testing, the results revealed SMOTE as the most effective for handling imbalanced data.
Performance of Sampling Methods on Imbalanced Data: Comparative Analysis
Imbalanced data affects a range of machine learning applications including bioengineering. Sampling is the most common approach to deal with imbalanced data. In this study, the effectiveness of various popular sampling algorithms that are used to balance imbalanced data were compared. Concretely, five popular sampling algorithms were considered - synthetic minority oversampling technique, random oversampling, adaptive synthetic sampling, Gamma, and random undersampling - in dealing with imbalanced data. The performance of three machine learning algorithms was evaluated - random forest, support vector machines, and deep neural network - on ten imbalanced datasets that are artificially balanced. Through multiple testing, the results revealed SMOTE as the most effective for handling imbalanced data.
Performance of Sampling Methods on Imbalanced Data: Comparative Analysis
Butt, Abdul Hadi (author) / Khan, Zahab (author) / Khan, Afreen (author) / Ghazanfar, Hanifa (author) / Zgheib, Rita (author) / Kamalov, Firuz (author)
2024-06-03
369510 byte
Conference paper
Electronic Resource
English
Work Sampling Programs: Comparative Analysis
| ASCE | 2021
Keep it simple: random oversampling for imbalanced data
| IEEE | 2023
Improving activated sludge classification based on imbalanced data
| British Library Online Contents | 2014
| DOAJ | 2023