ProSiT! Latent Variable Discovery with PROgressive SImilarity Thresholds

TL;DR

ProSiT is a deterministic, flexible method for discovering latent document dimensions that automatically determines the optimal number of topics, outperforming traditional topic models and clustering methods on multiple metrics.

Contribution

ProSiT introduces a novel, interpretable approach that finds the number of latent dimensions without stochasticity, requiring only two hyper-parameters and demonstrating superior performance.

Findings

ProSiT matches or outperforms existing methods on coherence and distinctiveness.

It produces replicable, deterministic results across benchmark datasets.

The method is agnostic to input format and easy to tune.

Abstract

The most common ways to explore latent document dimensions are topic models and clustering methods. However, topic models have several drawbacks: e.g., they require us to choose the number of latent dimensions a priori, and the results are stochastic. Most clustering methods have the same issues and lack flexibility in various ways, such as not accounting for the influence of different topics on single documents, forcing word-descriptors to belong to a single topic (hard-clustering) or necessarily relying on word representations. We propose PROgressive SImilarity Thresholds - ProSiT, a deterministic and interpretable method, agnostic to the input format, that finds the optimal number of latent dimensions and only has two hyper-parameters, which can be set efficiently via grid search. We compare this method with a wide range of topic models and clustering methods on four benchmark data sets. In most setting, ProSiT matches or outperforms the other methods in terms six metrics of topic coherence and distinctiveness, producing replicable, deterministic results.