International Conference on Computer and Knowledge Engineering

Home / 11th International Conference on Computer and Knowledge Engineering

A Novel Method For Fake News Detection Based on Propagation Tree

Authors :

Mansour Davoudi¹ Mohammad Reza Moosavi² Mohammad Hadi Sadreddini³

1- Department of Computer Science and Engineering and IT, School of Electrical and Computer Engineering, Shiraz University, Shiraz, Iran. 2- Department of Computer Science and Engineering and IT, School of Electrical and Computer Engineering, Shiraz University, Shiraz, Iran. 3- Department of Computer Science and Engineering and IT, School of Electrical and Computer Engineering, Shiraz University, Shiraz, Iran.

Keywords :

Fake news detection, Social media, LSTM network, Propagation tree

Abstract :

Nowadays, online social media play a significant role in news broadcast due to their convenience, speed, and accessibility. Social media platforms leverage the rapid production of a large volume of information and cause the propagation of untrustworthy and fake news. Since fake news is engineered to persuade a wide range of readers intentionally, it is difficult to detect them just based on the news content, and more information, such as the social context, is needed. In this paper, we propose a new model based on analyzing the propagation tree for detecting fake news. We dynamically extract novel features from the constructed propagation trees over time. To predict the veracity of a news article, a kind of recurrent neural network (LSTM) is used to capture the temporal dynamics of extracted features and identify the evolution pattern of the propagation tree over time. Our proposed model is evaluated on the FakeNewsNet repository, which consists of two recent well-known datasets in the field, namely PolitiFact and GossipCop. Our results show encouraging performance, outperforming the state-of-the-art methods by 2.3% on the PolitiFact and 1.2% on the GossipCop datasets