The Dynamic Embedded Topic Model

TL;DR

The paper introduces the dynamic embedded topic model (D-ETM), a novel approach combining word embeddings and dynamic topic modeling to better capture evolving themes in sequential document collections.

Contribution

It develops a new generative model that integrates word embeddings with dynamic latent Dirichlet allocation, enabling smooth, interpretable topic trajectories over time.

Findings

D-ETM outperforms D-LDA on document completion tasks.

D-ETM learns more diverse and coherent topics.

D-ETM requires less training time.

Abstract

Topic modeling analyzes documents to learn meaningful patterns of words. For documents collected in sequence, dynamic topic models capture how these patterns vary over time. We develop the dynamic embedded topic model (D-ETM), a generative model of documents that combines dynamic latent Dirichlet allocation (D-LDA) and word embeddings. The D-ETM models each word with a categorical distribution parameterized by the inner product between the word embedding and a per-time-step embedding representation of its assigned topic. The D-ETM learns smooth topic trajectories by defining a random walk prior over the embedding representations of the topics. We fit the D-ETM using structured amortized variational inference with a recurrent neural network. On three different corpora---a collection of United Nations debates, a set of ACL abstracts, and a dataset of Science Magazine articles---we found that the D-ETM outperforms D-LDA on a document completion task. We further found that the D-ETM learns more diverse and coherent topics than D-LDA while requiring significantly less time to fit.