Fast, Flexible Models for Discovering Topic Correlation across Weakly-Related Collections

TL;DR

This paper introduces two probabilistic models, C-LDA and C-HDP, designed to efficiently discover weakly-related topic correlations across large, asymmetric document collections, addressing limitations of existing models.

Contribution

The paper presents novel probabilistic models and a parallel sampling algorithm for large-scale, weakly-related collection analysis, improving detection of tail-end topic correlations.

Findings

Models effectively identify weak topic correlations in synthetic data.

Successful application to over 300,000 documents from JSTOR.

Parallel sampling enhances scalability and efficiency.

Abstract

Weak topic correlation across document collections with different numbers of topics in individual collections presents challenges for existing cross-collection topic models. This paper introduces two probabilistic topic models, Correlated LDA (C-LDA) and Correlated HDP (C-HDP). These address problems that can arise when analyzing large, asymmetric, and potentially weakly-related collections. Topic correlations in weakly-related collections typically lie in the tail of the topic distribution, where they would be overlooked by models unable to fit large numbers of topics. To efficiently model this long tail for large-scale analysis, our models implement a parallel sampling algorithm based on the Metropolis-Hastings and alias methods (Yuan et al., 2015). The models are first evaluated on synthetic data, generated to simulate various collection-level asymmetries. We then present a case study of modeling over 300k documents in collections of sciences and humanities research from JSTOR.