TL;DR
The MLIDSC introduces an automated labeling mechanism that eliminates the need for prior parameter assumptions and uses a novel weighted scheme that combines the imbalance ratio and the importance of individual instances at a given time, ensuring a focus on critical data points.
The results demonstrate that CADEE can improve label utilization, drift adaptation, and minority class recognition in non-stationary multi-class imbalanced data streams.
This work proposes DMAE, a dual-memory active ensemble learning method for multiclass imbalanced concept-drifting data streams, and proposes a composite sample-weighting formulation, PCN-Weight, which jointly models boundary difficulty, class-imbalance status, sample–prototype relations, and temporal decay to guide inc...
A new classification model called Feature-based Reservoir of Dynamic Subspaces (FeRDS) is proposed, which delivers superior prediction performance across various datasets and latency levels while maintaining low memory consumption and processing time.
A novel approach that leverages statistical redistribution in the output space to approximate the post-removal confidence vectors of a retrained model, alleviating scalability limitations and potentially mitigates privacy concerns inherent to data-dependent solutions is proposed.
The proposed iLEAD method combines hybrid genetic programming with a CART classifier, offering improved predictive performance, reduced processing time, and lower memory usage compared to traditional GP-based decision trees, making it ideal for incremental learning in dynamic datasets.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
