Stick-Breaking Embedded Topic Model with Continuous Optimal Transport for Online Analysis of Document Streams

TL;DR

SB-SETM is an online topic model that dynamically infers the number of active topics and merges topic embeddings using optimal transport, effectively analyzing evolving document streams.

Contribution

The paper introduces SB-SETM, a novel online topic model that combines a stick-breaking process with optimal transport for real-time document stream analysis.

Findings

Outperforms baseline models on simulated data

Effectively captures evolving topics in news articles

Automatically infers the number of active topics

Abstract

Online topic models are unsupervised algorithms to identify latent topics in data streams that continuously evolve over time. Although these methods naturally align with real-world scenarios, they have received considerably less attention from the community compared to their offline counterparts, due to specific additional challenges. To tackle these issues, we present SB-SETM, an innovative model extending the Embedded Topic Model (ETM) to process data streams by merging models formed on successive partial document batches. To this end, SB-SETM (i) leverages a truncated stick-breaking construction for the topic-per-document distribution, enabling the model to automatically infer from the data the appropriate number of active topics at each timestep; and (ii) introduces a merging strategy for topic embeddings based on a continuous formulation of optimal transport adapted to the high dimensionality of the latent topic space. Numerical experiments show SB-SETM outperforming baselines on simulated scenarios. We extensively test it on a real-world corpus of news articles covering the Russian-Ukrainian war throughout 2022-2023.