DEAP-FAKED: Knowledge Graph based Approach for Fake News Detection

Abstract

Fake News on social media platforms has attracted a lot of attention in recent times, primarily for events related to politics (2016 US Presidential elections), healthcare (infodemic during COVID-19), to name a few. Various methods have been proposed for detecting Fake News. The approaches span from exploiting techniques related to network analysis, Natural Language Processing (NLP), and the usage of Graph Neural Networks (GNNs). In this work, we propose DEAP-FAKED, a knowleDgE grAPh FAKe nEws Detection framework for identifying Fake News. Our approach is a combination of the NLP -where we encode the news content, and the GNN technique -where we encode the Knowledge Graph (KG). A variety of these encodings provides a complementary advantage to our detector. We evaluate our framework using two publicly available datasets containing articles from domains such as politics, business, technology, and healthcare. As part of dataset pre-processing, we also remove the bias, such as the source of the articles, which could impact the performance of the models. DEAP-FAKED obtains an F1-score of 88% and 78% for the two datasets, which is an improvement of ∼21%, and ∼3% respectively, which shows the effectiveness of the approach.

MCWDST: A Minimum-Cost Weighted Directed Spanning Tree Algorithm for Real-Time Fake News Mitigation in Social Media

“…This is mitigated in large datasets by the texts' length, which offers sufficient semantic, syntactic, and contextual information to train accurate and data-specific word embeddings. The proposed novel deep learning architectures offer stateof-the-art results on all tested datasets, i.e., an accuracy of 88.66 on Kaggle [25], 74.80 on Fakeddit [30], and 72.30 on GossipCop [33]. We note that the longer texts enable the bidirectional LSTM layers with and without a CNN layer on top to detect meaningful patterns for fake news detection.…”

“…Mayank et al [25] propose DEAP-FAKED, a new model that uses Natural Language Processing techniques, Graph Neural Networks, and Knowledge graphs to identify Fake News. The experimental results on the Kaggle dataset [26] show this approach improves the performance of misinformation detection.…”

The Study of Online Public Opinion Method Based on Emotional Analysis and CNN

“…Qian et al [48] added the Knowledge graph module to learn the external knowledge contained in news text entities under the traditional multimodal framework; However, it takes a lot of time to construct knowledge map training on the whole news text. In response to this defect, Mayank et al [49] proposed a false news detection framework that only targets Headline entities. Under this framework, they use two-way Long short-term memory (LSTM) networks to learn Headline features, and identify and extract Headline entities, build a Knowledge graph to obtain external information features, and finally splice and fuse the two parts of features to do the second classification task.…”

A Comprehensive Survey of Graph Neural Networks for Knowledge Graphs

“…The DEAP-FAKED [76] encoded news content using an NLP-based approach and then identified, extracted, and mapped the named entities to a KG. Finally, the entities in the KG were encoded using a GNN-based technique.…”