A Boundary-Aware Non-parametric Granular-Ball Classifier Based on Minimum Description Length
Researchers have introduced a new granular-ball classifier that uses the Minimum Description Length (MDL) principle to improve transparency and boundary sensitivity. This MDL-based Granular-Ball Classifier (MDL-GBC) formulates the construction of granular balls as a local model selection problem, comparing single-ball, two-ball, and core-boundary models. Experiments on 18 benchmark datasets demonstrate that MDL-GBC achieves competitive performance, often outperforming existing methods in accuracy and Macro-F1 scores, offering an interpretable alternative to traditional heuristic approaches. AI
IMPACT Introduces a more interpretable and boundary-aware classification method, potentially improving performance in specific machine learning tasks.