Can I read Clustering with Boosted Classifiers on EtoBox?

Clustering with Boosted Classifiers by aegr82 is a document available to read on EtoBox.

What is Clustering with Boosted Classifiers about?

The document presents CLOOSTING, a novel clustering approach that utilizes boosting techniques, specifically Adaboost, in an unsupervised context to iteratively model clusters. The method enhances within-cluster homogeneity and inter-cluster separation through outlier rejection, localized weak learners, and Gaussian kernel smoothing. Experimental results on various datasets demonstrate that CLOOSTING outperforms existing state-of-the-art methods, particularly in higher-dimensional spaces.

Author

aegr82

Language

EN