IMOVNO+: A Regional Partitioning and Meta-Heuristic Ensemble Framework for Imbalanced Multi-Class Learning
#IMOVNO+#Class Imbalance#Multi-Class Learning#Machine Learning#Data Quality#Ensemble Methods#Meta-Heuristic#Conditional Probability
📌 Key Takeaways
IMOVNO+ is a two-level framework addressing class imbalance, overlap, and noise in machine learning
The framework operates at both data and algorithmic levels with innovative techniques for each
Testing showed significant performance improvements over existing methods across 35 datasets
The approach particularly excels with multi-class data and in scenarios with data scarcity and privacy constraints
📖 Full Retelling
Researchers Soufiane Bacha, Laouni Djafri, Sahraoui Dhelim, and Huansheng Ning introduced IMOVNO+, a novel two-level framework for addressing class imbalance, overlap, and noise in machine learning, through a paper submitted to arXiv on February 22, 2026. The framework was developed to enhance data quality and algorithmic robustness for both binary and multi-class tasks, addressing persistent challenges that degrade model performance and limit generalization in complex datasets where traditional approaches struggle to capture distribution shapes and minority-majority relationships. IMOVNO+ operates at two levels: data and algorithmic. At the data level, it uses conditional probability to quantify sample informativeness, partitions datasets into core, overlapping, and noisy regions, implements an overlapping-cleaning algorithm combining Z-score metrics with big-jump gap distance, and employs a smart oversampling technique based on multi-regularization to control synthetic sample proximity. At the algorithmic level, a meta-heuristic approach prunes ensemble classifiers to reduce the influence of weak learners, addressing limitations in existing ensemble methods that struggle to integrate diverse classifiers effectively. The framework was evaluated on 35 datasets (13 multi-class, 22 binary) and demonstrated consistent superiority over state-of-the-art methods, approaching 100% performance in several cases. For multi-class data, IMOVNO+ achieved gains of 37-57% in G-mean, 25-44% in F1-score, 25-39% in precision, and 26-43% in recall. In binary tasks, it attained near-perfect performance with improvements of 14-39%. The researchers note that the framework particularly excels in handling data scarcity and imbalance arising from collection limitations and privacy constraints, making it valuable for real-world applications where such challenges are common.
🏷️ Themes
Machine Learning, Data Quality, Algorithm Optimization
In statistics and machine learning, ensemble methods use multiple learning algorithms to obtain better predictive performance than could be obtained from any of the constituent learning algorithms alone.
Unlike a statistical ensemble in statistical mechanics, which is usually infinite, a machine lea...
Probability of an event occurring, given that another event has already occurred
In probability theory, conditional probability is a measure of the probability of an event occurring, given that another event (by assumption, presumption, assertion, or evidence) is already known to have occurred. This particular method relies on event A occurring with some sort of relationship wit...
State of qualitative or quantitative pieces of information
Data quality refers to the state of qualitative or quantitative pieces of information. There are many definitions of data quality, but data is generally considered high quality if it is "fit for [its] intended uses in operations, decision making and planning". Data is deemed of high quality if it co...
Study of algorithms that improve automatically through experience
Machine learning (ML) is a field of study in artificial intelligence concerned with the development and study of statistical algorithms that can learn from data and generalize to unseen data, and thus perform tasks without explicit instructions. Within a subdiscipline in machine learning, advances i...
No entity connections available yet for this article.
Original Source
--> Computer Science > Machine Learning arXiv:2602.20199 [Submitted on 22 Feb 2026] Title: IMOVNO+: A Regional Partitioning and Meta-Heuristic Ensemble Framework for Imbalanced Multi-Class Learning Authors: Soufiane Bacha , Laouni Djafri , Sahraoui Dhelim , Huansheng Ning View a PDF of the paper titled IMOVNO+: A Regional Partitioning and Meta-Heuristic Ensemble Framework for Imbalanced Multi-Class Learning, by Soufiane Bacha and 3 other authors View PDF HTML Abstract: Class imbalance, overlap, and noise degrade data quality, reduce model reliability, and limit generalization. Although widely studied in binary classification, these issues remain underexplored in multi-class settings, where complex inter-class relationships make minority-majority structures unclear and traditional clustering fails to capture distribution shape. Approaches that rely only on geometric distances risk removing informative samples and generating low-quality synthetic data, while binarization approaches treat imbalance locally and ignore global inter-class dependencies. At the algorithmic level, ensembles struggle to integrate weak classifiers, leading to limited robustness. This paper proposes IMOVNO+ (IMbalance-OVerlap-NOise+ Algorithm-Level Optimization), a two-level framework designed to jointly enhance data quality and algorithmic robustness for binary and multi-class tasks. At the data level, first, conditional probability is used to quantify the informativeness of each sample. Second, the dataset is partitioned into core, overlapping, and noisy regions. Third, an overlapping-cleaning algorithm is introduced that combines Z-score metrics with a big-jump gap distance. Fourth, a smart oversampling algorithm based on multi-regularization controls synthetic sample proximity, preventing new overlaps. At the algorithmic level, a meta-heuristic prunes ensemble classifiers to reduce weak-learner influence. IMOVNO+ was evaluated on 35 datasets (13 multi-class, 22 binary). Results show consist...
