CRAFT: ClusteR-Specific Assorted Feature selecTion
Abstract
We present a framework for clustering with cluster-specific feature selection. The framework, CRAFT, is derived from asymptotic log posterior formulations of nonparametric MAP-based clustering models. CRAFT handles assorted data, i.e., both numeric and categorical data, and the underlying objective functions are intuitively appealing. The resulting algorithm is simple to implement and scales nicely, requires minimal parameter tuning, obviates the need to specify the number of clusters a priori, and compares favorably with other methods on real datasets.
Cite
Text
Garg et al. "CRAFT: ClusteR-Specific Assorted Feature selecTion." International Conference on Artificial Intelligence and Statistics, 2016.Markdown
[Garg et al. "CRAFT: ClusteR-Specific Assorted Feature selecTion." International Conference on Artificial Intelligence and Statistics, 2016.](https://mlanthology.org/aistats/2016/garg2016aistats-craft/)BibTeX
@inproceedings{garg2016aistats-craft,
title = {{CRAFT: ClusteR-Specific Assorted Feature selecTion}},
author = {Garg, Vikas K. and Rudin, Cynthia and Jaakkola, Tommi S.},
booktitle = {International Conference on Artificial Intelligence and Statistics},
year = {2016},
pages = {305-313},
url = {https://mlanthology.org/aistats/2016/garg2016aistats-craft/}
}