RP-DNN: A Tweet level propagation context based deep neural networks for early rumor detection in Social Media

TL;DR

This paper introduces RP-DNN, a novel deep neural network architecture that leverages tweet-level propagation context to detect rumors early in social media, outperforming existing methods especially in the initial stages of rumor spread.

Contribution

The paper presents a hybrid neural network combining character-based language models and LSTMs with attention mechanisms for early rumor detection at the message level.

Findings

Achieved state-of-the-art performance on seven real-life rumor datasets.

Effectively detects unseen rumors with large augmented data.

Component ablation highlights the importance of each model part.

Abstract

Early rumor detection (ERD) on social media platform is very challenging when limited, incomplete and noisy information is available. Most of the existing methods have largely worked on event-level detection that requires the collection of posts relevant to a specific event and relied only on user-generated content. They are not appropriate to detect rumor sources in the very early stages, before an event unfolds and becomes widespread. In this paper, we address the task of ERD at the message level. We present a novel hybrid neural network architecture, which combines a task-specific character-based bidirectional language model and stacked Long Short-Term Memory (LSTM) networks to represent textual contents and social-temporal contexts of input source tweets, for modelling propagation patterns of rumors in the early stages of their development. We apply multi-layered attention models to jointly learn attentive context embeddings over multiple context inputs. Our experiments employ a stringent leave-one-out cross-validation (LOO-CV) evaluation setup on seven publicly available real-life rumor event data sets. Our models achieve state-of-the-art(SoA) performance for detecting unseen rumors on large augmented data which covers more than 12 events and 2,967 rumors. An ablation study is conducted to understand the relative contribution of each component of our proposed model.