PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Evaluation methods for topic models
Hanna M Wallach, Iain Murray, Ruslan Salakhutdinov and David Mimno
Proceedings of the 26th International Conference on Machine Learning (ICML) 2009.

Abstract

A natural evaluation metric for statistical topic models is the probability of held-out documents given a trained model. While exact computation of this probability is intractable, several estimators for this probability have been used in the topic modeling literature, including the harmonic mean method and empirical likelihood method. In this paper, we demonstrate experimentally that commonly-used methods are unlikely to accurately estimate the probability of held-out documents, and propose two alternative methods that are both accurate and efficient.

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Natural Language Processing
Theory & Algorithms
ID Code:5939
Deposited By:Iain Murray
Deposited On:08 March 2010