A Discriminative Latent Variable Model for Online Clustering

This paper presents a latent variable structured prediction model for discriminative supervised clustering of items called the Latent Left-linking Model ((LM)-M-3). We present an online clustering algorithm for (LM)-M-3 based on a feature-based item similarity function. We provide a learning framework for estimating the similarity function and present a fast stochastic gradient-based learning technique. In our experiments on coreference resolution and document clustering, (LM)-M-3 outperforms several existing online as well as batch supervised clustering techniques.