New DK-GBMKKM algorithm adapts clustering to evolving kernel geometry

Researchers have developed a new clustering algorithm called DK-GBMKKM, which dynamically generates granular balls within the fused kernel space. This approach allows the representation to adapt to evolving kernel geometries during multiple kernel learning, unlike previous methods that generated balls once in the input space. The algorithm also incorporates a sample-size-weighted granular-ball kernel to maintain contributions from balls of varying sizes. Experiments on 12 datasets show DK-GBMKKM achieves strong clustering performance, and the code has been open-sourced. AI

IMPACT Introduces a novel clustering algorithm that adapts to evolving data geometries, potentially improving performance in machine learning tasks.

RANK_REASON This is a research paper detailing a novel algorithm. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• DK-GBMKKM
• Granular-ball-based Fast Spectral Embedding Clustering Algorithm for Large-Scale Data
• Hugging Face
• IArxiv
• Multiple Kernel k-Means with Incomplete Kernels

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →