Tag
The paper proposes a decision framework predicting when streaming oversampling adds value over continuous re-evaluation for imbalanced data streams with delayed labels and concept drift, evaluated across 15 classifiers on synthetic and real streams.
This paper proposes pretraining the CABiGRU temporal architecture on diffusion-generated sensor windows, then fine-tuning on real DEO smartwatch data, to improve recognition of underrepresented eating/drinking activities, achieving 90.6% balanced accuracy. The approach shows that diffusion-based synthetic data better preserves multivariate temporal structure than SMOTE/GAN augmentation for handling class imbalance in human activity recognition.
D-FROST introduces a decentralized federated prompt-tuning algorithm using optimal transport to address challenges with non-IID and imbalanced data, ensuring convergence and effectiveness for foundation models.
This study develops a bankruptcy prediction framework using hybrid resampling, stacking ensembles, and explainable AI to enhance minority-class detection in imbalanced financial data. The results highlight GRU with SMOTE-ENN as the best performer and SHAP analysis for identifying key bankruptcy risk predictors.
This paper studies the application of AI and machine learning algorithms for recognizing fraudulent banking transactions, proposing preprocessing techniques and comparing models. An artificial neural network and stacked generalization achieve improved AUC scores, with the best result around 0.954.
This paper investigates deep clustering methods on imbalanced tabular data and proposes two novel ensemble approaches that aggregate clustering assignments across embedding dimensions or via majority voting, outperforming individual methods on 16 datasets.
The paper introduces a dependency-aware fidelity diagnostic to measure inter-column dependency in synthetic tabular data, revealing that standard metrics are blind to dependency and that current generators have a residual gap not closed by capacity increases or common fixes.
Introduces K-IPO, a generate-then-select oversampling framework that preserves the original data's feature importance ranking (measured by Kendall's tau) during augmentation for imbalanced tabular data, showing improved preservation, explanation consistency, and predictive performance across 20 datasets.