Evaluating Dynamic Topic Models

TL;DR

This paper introduces a new quantitative evaluation measure for dynamic topic models that assesses topic quality over time and correlates well with human judgment, aiding in model comparison and analysis.

Contribution

The paper proposes a novel evaluation measure for DTMs that combines topic quality and temporal consistency, filling a key gap in model assessment.

Findings

The measure correlates well with human judgment.

It effectively identifies changing topics over time.

Demonstrated utility on synthetic and real data.

Abstract

There is a lack of quantitative measures to evaluate the progression of topics through time in dynamic topic models (DTMs). Filling this gap, we propose a novel evaluation measure for DTMs that analyzes the changes in the quality of each topic over time. Additionally, we propose an extension combining topic quality with the model's temporal consistency. We demonstrate the utility of the proposed measure by applying it to synthetic data and data from existing DTMs. We also conducted a human evaluation, which indicates that the proposed measure correlates well with human judgment. Our findings may help in identifying changing topics, evaluating different DTMs, and guiding future research in this area.