Experimental Evaluation of Dynamic Topic Modeling Algorithms

TL;DR

This paper compares various dynamic topic modeling algorithms using a new assessment metric to evaluate how effectively they capture topic evolution over time in large social media datasets.

Contribution

It provides a comprehensive quantitative comparison of dynamic topic models and introduces a novel metric for assessing topic change over time.

Findings

Dynamic models vary significantly in capturing topic evolution.

The proposed metric effectively measures temporal topic changes.

Some models outperform others in stability and adaptability.

Abstract

The amount of text generated daily on social media is gigantic and analyzing this text is useful for many purposes. To understand what lies beneath a huge amount of text, we need dependable and effective computing techniques from self-powered topic models. Nevertheless, there are currently relatively few thorough quantitative comparisons between these models. In this study, we compare these models and propose an assessment metric that documents how the topics change in time.