Publication:
Multimodal networks

Placeholder

Program

KU-Authors

KU Authors

Co-Authors

Mark, James E

Advisor

Publication Date

Language

English

Journal Title

Journal ISSN

Volume Title

Abstract

The real world involves many graphs and networks that are essentially heterogeneous, in which various types of relations connect multiple types of vertices. With the development of information networks, node features can be described by data of different modalities, resulting in multimodal heterogeneous graphs. However, most existed methods can only handle unimodal heterogeneous graphs. Moreover, most existing heterogeneous graph mining methods are based on meta-paths that depend on domain experts for modeling. In this paper, we propose a novel multimodal heterogeneous graph attention network (MHGAT) to address these problems. Specifically, we exploit edge-level aggregation to capture graph heterogeneity information to achieve more informative representations adaptively. Further, we use the modality-level attention mechanism to obtain multimodal fusion information. Because plain graph convolutional networks can not capture higher-order neighborhood information, we utilize the residual connection and the dense connection access to obtain it. Extensive experimental results show that the MHGAT outperforms state-of-the-art baselines on three datasets for node classification, clustering, and visualization tasks.

Description

Source:

Rubberlike Elasticity: A Molecular Primer, 2nd Edition

Publisher:

Cambridge Univ Press

Keywords:

Subject

Chemistry, Applied chemistry, Polymer science

Citation

Endorsement

Review

Supplemented By

Referenced By

Copy Rights Note

0

Views

0

Downloads

View PlumX Details