# Choosing the Number of Topics in LDA Models -- A Monte Carlo Comparison of Selection Criteria

arXiv: https://arxiv.org/abs/2212.14074 · HTML: https://mlsanity.com/paper/2212.14074/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 24 references that are in the map.
- First posted on arXiv 28 Dec 2022.

## Builds on (most cited first, 4 of 4)

- [A Bayesian information criterion for singular models](https://mlsanity.com/paper/1309.0911/) (2013)
- [Measuring Emotions in the COVID-19 Real World Worry Dataset](https://mlsanity.com/paper/2004.04225/) (2020)
- [The Exact Asymptotic Form of Bayesian Generalization Error in Latent Dirichlet Allocation](https://mlsanity.com/paper/2008.01304/) (2020)
- [Asymptotic Bayesian Generalization Error in Latent Dirichlet Allocation and Stochastic Matrix Factorization](https://mlsanity.com/paper/1709.04212/) (2017)

## Built on by (most cited first, 3 of 3)

- [Topic-aware Large Language Models for Summarizing the Lived Healthcare Experiences Described in Health Stories](https://mlsanity.com/paper/2510.24765/) (2025)
- [The Confounder Trap: Treatment-Encoding Representations in Causal Inference with Text](https://mlsanity.com/paper/2607.26309/) (2026)
- [Tipping Points in LLM-Based Multi-Agent Systems: Stance on Climate Change Action](https://mlsanity.com/paper/2609.25432/) (2026)

## Abstract (arXiv)

Selecting the number of topics in LDA models is considered to be a difficult task, for which alternative approaches have been proposed. The performance of the recently developed singular Bayesian information criterion (sBIC) is evaluated and compared to the performance of alternative model selection criteria. The sBIC is a generalization of the standard BIC that can be implemented to singular statistical models. The comparison is based on Monte Carlo simulations and carried out for several alternative settings, varying with respect to the number of topics, the number of documents and the size of documents in the corpora. Performance is measured using different criteria which take into account the correct number of topics, but also whether the relevant topics from the DGPs are identified. Practical recommendations for LDA model selection in applications are derived.