Enhancing Temporal Link Prediction with HierTKG: A Hierarchical Temporal Knowledge Graph Framework

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Almutairi, Mariam, Aktas, Melike Yildiz, Wali, Nawar, Mitra, Shutonu, Zhou, Dawei
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909430840295424
author Almutairi, Mariam
Aktas, Melike Yildiz
Wali, Nawar
Mitra, Shutonu
Zhou, Dawei
author_facet Almutairi, Mariam
Aktas, Melike Yildiz
Wali, Nawar
Mitra, Shutonu
Zhou, Dawei
contents The rapid spread of misinformation on social media, especially during crises, challenges public decision-making. To address this, we propose HierTKG, a framework combining Temporal Graph Networks (TGN) and hierarchical pooling (DiffPool) to model rumor dynamics across temporal and structural scales. HierTKG captures key propagation phases, enabling improved temporal link prediction and actionable insights for misinformation control. Experiments demonstrate its effectiveness, achieving an MRR of 0.9845 on ICEWS14 and 0.9312 on WikiData, with competitive performance on noisy datasets like PHEME (MRR: 0.8802). By modeling structured event sequences and dynamic social interactions, HierTKG adapts to diverse propagation patterns, offering a scalable and robust solution for real-time analysis and prediction of rumor spread, aiding proactive intervention strategies.
format Preprint
id arxiv_https___arxiv_org_abs_2412_12385
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Enhancing Temporal Link Prediction with HierTKG: A Hierarchical Temporal Knowledge Graph Framework
Almutairi, Mariam
Aktas, Melike Yildiz
Wali, Nawar
Mitra, Shutonu
Zhou, Dawei
Social and Information Networks
Artificial Intelligence
The rapid spread of misinformation on social media, especially during crises, challenges public decision-making. To address this, we propose HierTKG, a framework combining Temporal Graph Networks (TGN) and hierarchical pooling (DiffPool) to model rumor dynamics across temporal and structural scales. HierTKG captures key propagation phases, enabling improved temporal link prediction and actionable insights for misinformation control. Experiments demonstrate its effectiveness, achieving an MRR of 0.9845 on ICEWS14 and 0.9312 on WikiData, with competitive performance on noisy datasets like PHEME (MRR: 0.8802). By modeling structured event sequences and dynamic social interactions, HierTKG adapts to diverse propagation patterns, offering a scalable and robust solution for real-time analysis and prediction of rumor spread, aiding proactive intervention strategies.
title Enhancing Temporal Link Prediction with HierTKG: A Hierarchical Temporal Knowledge Graph Framework
topic Social and Information Networks
Artificial Intelligence
url https://arxiv.org/abs/2412.12385