Inference and Evaluation of the Multinomial Mixture Model for Text Clustering

TL;DR

This paper explores probabilistic multinomial mixture models for text clustering, comparing inference methods like EM and Gibbs sampling, and proposing evaluation criteria and heuristics to improve clustering accuracy in high-dimensional settings.

Contribution

It introduces a systematic evaluation framework for text clustering, compares EM and Gibbs sampling inference methods, and proposes heuristics for high-dimensional parameter estimation.

Findings

Gibbs sampling outperforms EM in high-dimensional scenarios.

Initialization and feature choices significantly affect clustering results.

Heuristic vocabulary reduction improves inference efficiency.

Abstract

In this article, we investigate the use of a probabilistic model for unsupervised clustering in text collections. Unsupervised clustering has become a basic module for many intelligent text processing applications, such as information retrieval, text classification or information extraction. The model considered in this contribution consists of a mixture of multinomial distributions over the word counts, each component corresponding to a different theme. We present and contrast various estimation procedures, which apply both in supervised and unsupervised contexts. In supervised learning, this work suggests a criterion for evaluating the posterior odds of new documents which is more statistically sound than the "naive Bayes" approach. In an unsupervised context, we propose measures to set up a systematic evaluation framework and start with examining the Expectation-Maximization (EM) algorithm as the basic tool for inference. We discuss the importance of initialization and the influence of other features such as the smoothing strategy or the size of the vocabulary, thereby illustrating the difficulties incurred by the high dimensionality of the parameter space. We also propose a heuristic algorithm based on iterative EM with vocabulary reduction to solve this problem. Using the fact that the latent variables can be analytically integrated out, we finally show that Gibbs sampling algorithm is tractable and compares favorably to the basic expectation maximization approach.