Changepoint Analysis of Topic Proportions in Temporal Text Data

TL;DR

This paper introduces a scalable method for detecting changepoints in large-scale textual data by modeling topic proportions over time, enabling automated and interpretable identification of shifts in topic popularity.

Contribution

We develop a novel computationally efficient approach for offline changepoint detection in large textual datasets using a specialized temporal topic model and likelihood ratio testing.

Findings

Successfully identified historically significant changepoints in literature and physics datasets.

Method detects both known and novel shifts in topic structures.

Provides interpretable insights into the evolution of topics over time.

Abstract

Changepoint analysis deals with unsupervised detection and/or estimation of time-points in time-series data, when the distribution generating the data changes. In this article, we consider \emph{offline} changepoint detection in the context of large scale textual data. We build a specialised temporal topic model with provisions for changepoints in the distribution of topic proportions. As full likelihood based inference in this model is computationally intractable, we develop a computationally tractable approximate inference procedure. More specifically, we use sample splitting to estimate topic polytopes first and then apply a likelihood ratio statistic together with a modified version of the wild binary segmentation algorithm of Fryzlewicz et al. (2014). Our methodology facilitates automated detection of structural changes in large corpora without the need of manual processing by domain experts. As changepoints under our model correspond to changes in topic structure, the estimated changepoints are often highly interpretable as marking the surge or decline in popularity of a fashionable topic. We apply our procedure on two large datasets: (i) a corpus of English literature from the period 1800-1922 (Underwoodet al., 2015); (ii) abstracts from the High Energy Physics arXiv repository (Clementet al., 2019). We obtain some historically well-known changepoints and discover some new ones.