Historia Magistra Vitae: Dynamic Topic Modeling of Roman Literature using Neural Embeddings
This work addresses the challenge of making dynamic topic modeling more accessible and interpretable for historical researchers, though it is incremental as it applies an existing neural method to a new domain.
The paper tackled the problem of dynamic topic modeling for historical analysis by comparing traditional statistical models (LDA and NMF) with a BERT-based neural model on Roman literature, finding that while quantitative metrics favored statistical models, the neural model provided better qualitative insights and was less sensitive to hyperparameters.
Dynamic topic models have been proposed as a tool for historical analysis, but traditional approaches have had limited usefulness, being difficult to configure, interpret, and evaluate. In this work, we experiment with a recent approach for dynamic topic modeling using BERT embeddings. We compare topic models built using traditional statistical models (LDA and NMF) and the BERT-based model, modeling topics over the entire surviving corpus of Roman literature. We find that while quantitative metrics prefer statistical models, qualitative evaluation finds better insights from the neural model. Furthermore, the neural topic model is less sensitive to hyperparameter configuration and thus may make dynamic topic modeling more viable for historical researchers.