Proceedings of the Fifth International Workshop on Learning with Imbalanced Domains: Theory and Applications Held in ECML-PKDD, Turin, Italy on 18 September 2023 Published as Volume 241 by the Proceedings of Machine Learning Research on 03 June 2024. Volume Edited by: Nuno Moniz Paula Branco Luis Torgo Nathalie Japkowicz Michal Wozniak Shuo Wang Series Editors: Neil D. Lawrence https://proceedings.mlr.press/v241/ Sun, 05 Jul 2026 14:57:46 +0000 Sun, 05 Jul 2026 14:57:46 +0000 Jekyll v3.10.0 Imbalanced data poses a significant challenge in classification as model performance is af- fected by insufficient learning from minority classes. Balancing methods are often used to address this problem. However, such techniques can lead to problems such as overfitting or loss of information. This study addresses a more challenging aspect of balancing methods - their impact on model behavior. To capture these changes, Explainable Artificial Intel- ligence tools are used to compare models trained on datasets before and after balancing. In addition to the Variable Importance method, this study uses Partial Dependence and Accumulated Local Effects profiles. Real and simulated datasets are tested, and an open- source Python package edgaro is developed to facilitate this analysis. The results obtained show significant changes in model behavior due to balancing methods, which can lead to biased models toward a balanced distribution. These findings confirm that balancing analysis should go beyond model performance comparisons to achieve higher reliability of machine learning models. Therefore, we propose a new method performance gain plot for informed data balancing strategy to make an optimal selection of balancing method by analyzing the measure of change in model behavior versus performance gain. Mon, 03 Jun 2024 00:00:00 +0000 https://proceedings.mlr.press/v241/stando24a.html https://proceedings.mlr.press/v241/stando24a.html Efficient management of imbalanced time series data is of paramount importance when data located in the tails, particularly extreme values, have a substantial influence on predic- tive outcomes. This paper introduces FSDA (Feature Selection and Data Augmentation), a combined approach of feature selection and data augmentation, to address this issue. FSDA aims to identify the most predictive features for tail data, which may exhibit differ- ent sensitivities compared to the rest of the dataset. Data augmentation, a conventional technique for handling imbalanced data, is employed to enhance the accuracy of machine learning regression methods. Augmented information is strategically incorporated using time-warping and drift methods to maintain the temporal integrity of the data. Empirical evidence based on a use case in financial data reveals that FSDA consistently outperforms feature selection (FS) and data augmentation (DA) methods across all percentiles ranging from 85 to 99, demonstrating its efficacy in managing imbalanced time series data and improving predictive accuracy. Mon, 03 Jun 2024 00:00:00 +0000 https://proceedings.mlr.press/v241/krief24a.html https://proceedings.mlr.press/v241/krief24a.html Improving the classification of multi-class imbalanced data is more difficult than its two- class counterpart. In this paper, we use deep neural networks to train new representations of tabular multi-class data. Unlike the typically developed re-sampling pre-processing meth- ods, our proposal modifies the distribution of features, i.e. the positions of examples in the learned embedded representation, and it does not modify the class sizes. To learn such embedded representations we introduced various definitions of triplet loss functions: the simplest one uses weights related to the degree of class imbalance, while the next pro- posals are intended for more complex distributions of examples and aim to generate a safe neighborhood of minority examples. Similarly to the resampling approaches, after applying such preprocessing, different classifiers can be trained on new representations. Experiments with popular multi-class imbalanced benchmark data sets and three classifiers showed the advantage of the proposed approach over popular pre-processing methods as well as basic versions of neural networks with classical loss function formulations. Mon, 03 Jun 2024 00:00:00 +0000 https://proceedings.mlr.press/v241/horna24a.html https://proceedings.mlr.press/v241/horna24a.html Learning classifiers from imbalanced data is known to be a challenging and important prob- lem in machine learning. As a results, the topic has been studied from a wide variety of angles. This includes the choice of evaluation measures and understanding the implica- tions of minority class subconcepts on model learning. In this work, however, we argue that the community may not be using precise enough evaluation measures when assessing the performance of imbalanced learning pipelines on data that includes an imbalance in the minority class subconcepts. We show that the performance estimates from standard measures used in imbalance learning are biased towards the largest minority subconcepts, and that standard imbalance correction techniques can exacerbate the bias. Finally, we demonstrate that the bias can, in part, be corrected by applying instance weighting in the evaluation measures. Mon, 03 Jun 2024 00:00:00 +0000 https://proceedings.mlr.press/v241/bellinger24a.html https://proceedings.mlr.press/v241/bellinger24a.html