Importance-aware Topic Modeling for Discovering Public Transit Risk from Noisy Social Media

TL;DR

This paper introduces an importance-aware topic modeling framework that leverages social media data to detect urban transit risks, effectively handling noisy, sparse signals by modeling linguistic interactions and user influence.

Contribution

It proposes a novel Poisson Deconvolution Factorization method that jointly models influence-weighted keywords and topics, improving interpretability and robustness in transit risk detection from social media.

Findings

Achieves state-of-the-art topic coherence on large social streams.

Demonstrates strong diversity and interpretability of learned topics.

Provides publicly available code and dataset for further research.

Abstract

Urban transit agencies increasingly turn to social media to monitor emerging service risks such as crowding, delays, and safety incidents, yet the signals of concern are sparse, short, and easily drowned by routine chatter. We address this challenge by jointly modeling linguistic interactions and user influence. First, we construct an influence-weighted keyword co-occurrence graph from cleaned posts so that socially impactful posts contributes proportionally to the underlying evidence. The core of our framework is a Poisson Deconvolution Factorization (PDF) that decomposes this graph into a low-rank topical structure and topic-localized residual interactions, producing an interpretable topic--keyword basis together with topic importance scores. A decorrelation regularizer \emph{promotes} distinct topics, and a lightweight optimization procedure ensures stable convergence under nonnegativity and normalization constraints. Finally, the number of topics is selected through a coherence-driven sweep that evaluates the quality and distinctness of the learned topics. On large-scale social streams, the proposed model achieves state-of-the-art topic coherence and strong diversity compared with leading baselines. The code and dataset are publicly available at https://github.com/pangjunbiao/Topic-Modeling_ITS.git