Publication Type

Conference Proceeding Article

Version

acceptedVersion

Publication Date

7-2024

Abstract

Texts are often interconnected in a network structure, e.g., academic papers via citations. On the one hand, though Graph Neural Networks (GNNs) have shown promising ability to derive effective embeddings for networked documents, they do not assume latent topics, resulting in uninterpretahle embeddings. On the other hand, topic models can infer interpretable document representations. However, most topic models focus on plain text and fail to leverage network structure across documents. In this paper, we propose a GNN-based topic model that both captures network connection and derives semantically interpretable text representations. For network modeling, we build our model with Optimal Transport Barycenter. For semantic interpretability, we extend optimal transport with pre-trained word embeddings.

Keywords

Graph Neural Networks, Text Mining, Optimal Transport, Dirichlet Distribution, Document Networks

Discipline

Artificial Intelligence and Robotics | Computer Sciences

Research Areas

Data Science and Engineering; Intelligent Systems and Optimization

Publication

Proceedings of the 40th IEEE International Conference on Data Engineering (ICDE 2024) : Utrecht, Netherlands, May 13-17

First Page

5743

Last Page

5744

Identifier

10.1109/ICDE60146.2024.00503

Publisher

IEEE

City or Country

Utrecht, Netherlands

Comments

PDF provided by faculty

Additional URL

https://doi.org/10.1109/ICDE60146.2024.00503

Download

Share

COinS