A Dual Imbalanced Classification Framework With Feature Transfer Guided by Memory Compensation Strategy
- 25 February 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 38 (5) , 2508-2524
Abstract
Fully mining the differential features of different class samples in overlapping areas is the key and difficult point to improving imbalanced classification performance under complex distribution patterns. Although existing data-level and algorithm-level methods have achieved good results in dealing with overlapping problems, sample generation and classifier training heavily rely on distribution information, and the ability to mine the different information is limited. This paper proposes a dual imbalanced classification framework with feature transfer guided by memory compensation strategy, which enhances the model's ability to mine differential features by constructing a feature space with better inter-class separability. In the traditional classification branch, a feature extraction network maps original samples to feature space and a traditional classifier is used to classify the features. In the compensation classification branch, a feature memory module based on iterative clustering strategy is designed, separately obtaining and saving the correctly classified feature centers of different classes. Moreover, a feature transfer module based on vector combination theory is proposed, combining “push” and “pull” vectors to transfer the misclassified features to the non-overlapping areas corresponding to the same class feature memory module, thereby constructing a feature space with better inter-class separability. Finally, a classification compensation strategy based on feature similarity is designed, integrating the prediction results of the traditional classifier and feature memory module as the final classification results. Experimental results on 50 imbalanced datasets show the proposed method outperforms 28 typical imbalanced classification methods in F1-score and G-mean. Especially on 20 severely overlapping datasets, the performance improvement is more significant.Keywords
Funding Information
- High-Performance Computing Platform of BUPT
This publication has 48 references indexed in Scilit:
- A survey on imbalanced learning: latest research, applications and future directionsArtificial Intelligence Review, 2024
- Hybrid density-based adaptive weighted collaborative representation for imbalanced learningApplied Intelligence, 2024
- Imbalanced complemented subspace representation with adaptive weight learningExpert Systems with Applications, 2024
- Imbalanced credit card fraud detection data: A solution based on hybrid neural network and clustering-based undersampling techniqueApplied Soft Computing, 2024
- WRND: A weighted oversampling framework with relative neighborhood density for imbalanced noisy classificationExpert Systems with Applications, 2023
- SMOTE-kTLNN: A hybrid re-sampling method based on SMOTE and a two-layer nearest neighbor classifierExpert Systems with Applications, 2023
- SWSEL: Sliding Window-based Selective Ensemble Learning for class-imbalance problemsEngineering Applications of Artificial Intelligence, 2023
- An ensemble imbalanced classification method based on model dynamic selection driven by data partition hybrid samplingExpert Systems with Applications, 2020
- Pattern recognition-based chillers fault detection method using Support Vector Data Description (SVDD)Applied Energy, 2013
- Borderline-SMOTE: A New Over-Sampling Method in Imbalanced Data Sets Learning2005