A non-parametric mixture model for topic modeling over time

TL;DR

This paper introduces npTOT, a non-parametric, flexible model for capturing evolving topics over time in large corpora, addressing limitations of previous models with limited temporal variation and high computational costs.

Contribution

The paper presents npTOT, a non-parametric model that allows an unbounded number of topics with flexible temporal dynamics, along with an efficient collapsed Gibbs sampler.

Findings

npTOT outperforms existing models on synthetic data.

The model effectively captures complex temporal topic variations.

Experimental results demonstrate improved modeling of long-term corpora.

Abstract

A single, stationary topic model such as latent Dirichlet allocation is inappropriate for modeling corpora that span long time periods, as the popularity of topics is likely to change over time. A number of models that incorporate time have been proposed, but in general they either exhibit limited forms of temporal variation, or require computationally expensive inference methods. In this paper we propose non-parametric Topics over Time (npTOT), a model for time-varying topics that allows an unbounded number of topics and exible distribution over the temporal variations in those topics' popularity. We develop a collapsed Gibbs sampler for the proposed model and compare against existing models on synthetic and real document sets.