Large margin classifiers to generate synthetic data for imbalanced datasets

Marques, Marcelo Ladeira; Villela, Saulo Moraes; Borges, Carlos Cristiano Hasenclever

doi:10.1007/s10489-020-01719-y

Cited by 6 publications

(3 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite the potential of ML methods, most of them are generally hampered by the class imbalance problem, which occurs when the proportion of samples of one class greatly outnumbers the others [4,5]. Since most ML algorithms are built to work with balanced datasets, the classifiers are biased toward the majority class.…”

Section: Of 22mentioning

confidence: 99%

“…Since most ML algorithms are built to work with balanced datasets, the classifiers are biased toward the majority class. To deal with this, several methods have been proposed in the literature [4][5][6][7], which can be classified into two types, algorithmic-level and data-level approaches. The former adapts the loss function of the algorithm by assigning a higher weight to the misclassification of samples associated with the minority classes during the training process [8].…”

Section: Of 22mentioning

confidence: 99%

See 1 more Smart Citation

Evaluation of Synthetic Categorical Data Generation Techniques for Predicting Cardiovascular Diseases and Post-hoc Interpretability of the Risk Factors

García-Vicente¹,

Chushig-Muzo²,

Jiménez³

et al. 2023

Preprint

View full text Add to dashboard Cite

Machine Learning (ML) methods have become important to enhance the performance of decision-support predictive models. However, class imbalance is one of the main challenges for developing ML models, because it limits the generalization of these models, and biases the learning algorithms. In this paper, we consider oversampling methods for generating synthetic categorical clinical data aiming to improve the predictive performance in ML models, and the identification of risk factors for cardiovascular diseases (CVDs). We performed a comparative study of several categorical synthetic data generation methods, including Generative Adversarial Networks (GANs). Then, we assessed the impact of combining oversampling strategies and linear and nonlinear supervised ML methods. Lastly, we conducted a post-hoc model interpretability based on the importance of the risk factors. Experimental results show the potential of GAN-based models for generating high-quality categorical synthetic data, yielding probability mass functions that are highly close to real data, maintaining relevant insights, and contributing to increase the predictive performance. The GAN-based model and a linear classifier outperforms other oversampling techniques, improving 2\% the area under the curve. These results demonstrate the capability of synthetic data to help both in determining risk factors and building models for CVD prediction.

show abstract

Section: Of 22mentioning

confidence: 99%

Section: Of 22mentioning

confidence: 99%

Evaluation of Synthetic Categorical Data Generation Techniques for Predicting Cardiovascular Diseases and Post-hoc Interpretability of the Risk Factors

García-Vicente¹,

Chushig-Muzo²,

Jiménez³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The adoption of synthetic data is becoming a feasible solution to this problem (Alloza et al, 2023;El Emam et al, 2020;Ladeira Marques et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

A Multi-Split Cross-Strategy for Enhancing Machine Learning Algorithms Prediction Results with Data Generated by Conditional Generative Adversarial Network

Abassi,

Bakkas,

Bakkas

et al. 2024

Journal of Computer Science

View full text Add to dashboard Cite

In this study, we present a Multi-Split Cross-Strategy (MSC-Strategy) designed to leverage synthetic tabular data generated by a Conditional Generative Adversarial Network (CGAN). Our study aims to investigate the potential of synthetic data in comparison to real-world data for improving machine learning predictive results. Firstly, we develop a CGAN architecture tailored to generate synthetic tabular data, trained on a comprehensive real-world dataset. Secondly, we validate the synthetic data generated by the CGAN to ensure its statistical fidelity and resemblance to the distribution of real data. Finally, we selectively leverage a subset of the generated data and apply our strategy to create a new combined training set comprising the training set of real data and the chosen subset of generated data. To validate our approach, we employ six diverse regression models: Decision Tree (DT), K-Nearest Neighbors (KNN), Random Forest (RF), XGB Regressor (XGB), and Support Vector Regressor (SVR). Each model is trained and tested using a training set of real data, generated data, combined data (training set of real data and generated data), and data formed by our MSC strategy. Our findings indicate that the training set formed by our MSC strategy demonstrates remarkable predictive performance compared to real-world data and generated data, highlighting its ability to enhance the prediction of machine learning models using only a subset of generated data.

show abstract

Borderline-margin loss based deep metric learning framework for imbalanced data

Yan

2022

Appl Intell

View full text Add to dashboard Cite

Large margin classifiers to generate synthetic data for imbalanced datasets

Cited by 6 publications

References 30 publications

Evaluation of Synthetic Categorical Data Generation Techniques for Predicting Cardiovascular Diseases and Post-hoc Interpretability of the Risk Factors

Evaluation of Synthetic Categorical Data Generation Techniques for Predicting Cardiovascular Diseases and Post-hoc Interpretability of the Risk Factors

A Multi-Split Cross-Strategy for Enhancing Machine Learning Algorithms Prediction Results with Data Generated by Conditional Generative Adversarial Network

Borderline-margin loss based deep metric learning framework for imbalanced data

Contact Info

Product

Resources

About