Topic Modeling on Health Journals with Regularized Variational Inference

TL;DR

This paper introduces the DAP model, a novel probabilistic approach for topic modeling in asynchronous, multi-author health journal datasets, utilizing regularized variational inference to improve topic distinctiveness and capture shared health journeys.

Contribution

The paper presents the DAP model, a new probabilistic graphical model for temporal, multi-author corpora, with a regularized inference algorithm to enhance persona distinction and topic modeling accuracy.

Findings

DAP outperforms existing models on health journal data

Regularization improves persona distinctiveness

Model captures shared health journeys among authors

Abstract

Topic modeling enables exploration and compact representation of a corpus. The CaringBridge (CB) dataset is a massive collection of journals written by patients and caregivers during a health crisis. Topic modeling on the CB dataset, however, is challenging due to the asynchronous nature of multiple authors writing about their health journeys. To overcome this challenge we introduce the Dynamic Author-Persona topic model (DAP), a probabilistic graphical model designed for temporal corpora with multiple authors. The novelty of the DAP model lies in its representation of authors by a persona --- where personas capture the propensity to write about certain topics over time. Further, we present a regularized variational inference algorithm, which we use to encourage the DAP model's personas to be distinct. Our results show significant improvements over competing topic models --- particularly after regularization, and highlight the DAP model's unique ability to capture common journeys shared by different authors.