IBM T.J. Watson Research; UC Berkeley; RPI & NSF

Unsupervised Feature Selection for the $k$-means Clustering Problem

7:00 – 11:59pm Monday, December 07, 2009

This is part of the Poster Session which begins at 19:00 on Monday December 7, 2009

M27

We present a novel feature selection algorithm for the $k$-means clustering problem. Our algorithm is randomized and, assuming an accuracy parameter $\epsilon \in (0,1)$, selects and appropriately rescales in an unsupervised manner $\Theta(k \log(k / \epsilon) / \epsilon^2)$ features from a dataset of arbitrary dimensions. We prove that, if we run any $\gamma$-approximate $k$-means algorithm ($\gamma \geq 1$) on the features selected using our method, we can find a $(1+(1+\epsilon)\gamma)$-approximate partition with high probability.

Download the PDF