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

arXiv: https://arxiv.org/abs/2512.06293 · HTML: https://mlsanity.com/paper/2512.06293/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 47 references that are in the map.
- First posted on arXiv 6 Dec 2025.

## Builds on (most cited first, 10 of 10)

- [Graph Convolutional Networks for Text Classification](https://mlsanity.com/paper/1809.05679/) (2018)
- [Topic Modeling in Embedding Spaces](https://mlsanity.com/paper/1907.04907/) (2019)
- [CleanNet: Transfer Learning for Scalable Image Classifier Training with Label Noise](https://mlsanity.com/paper/1711.07131/) (2017)
- [Discovering Discrete Latent Topics with Neural Variational Inference](https://mlsanity.com/paper/1706.00359/) (2017)
- [Topic Modelling Meets Deep Neural Networks: A Survey](https://mlsanity.com/paper/2103.00498/) (2021)
- [A Survey on Neural Topic Models: Methods, Applications, and Challenges](https://mlsanity.com/paper/2401.15351/) (2024)
- [Neural Topic Modeling by Incorporating Document Relationship Graph](https://mlsanity.com/paper/2009.13972/) (2020)
- [Concept Drift Adaptation in Text Stream Mining Settings: A Systematic Review](https://mlsanity.com/paper/2312.02901/) (2023)
- [Efficient Correlated Topic Modeling with Topic Embedding](https://mlsanity.com/paper/1707.00206/) (2017)
- [Finding A Taxi with Illegal Driver Substitution Activity via Behavior Modelings](https://mlsanity.com/paper/2404.11844/) (2024)

## Built on by (most cited first, 2 of 2)

- [Discovering Sparse Counterfactual Factors via Latent Adjustment for Survey-based Community Intervention](https://mlsanity.com/paper/2605.04460/) (2026)
- [Artificial Intelligence for Understanding and Managing Transportation Behavior in Sustainable Smart Cities](https://mlsanity.com/paper/2607.17694/) (2026)

## Abstract (arXiv)

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 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-ModelingITS.git