$$\mathsf {SAFE}$$: Similarity-Aware Multi-modal Fake News Detection

Abstract

Effective detection of fake news has recently attracted significant attention. Current studies have made significant contributions to predicting fake news with less focus on exploiting the relationship (similarity) between the textual and visual information in news articles. Attaching importance to such similarity helps identify fake news stories that, for example, attempt to use irrelevant images to attract readers' attention. In this work, we propose a Similarity-Aware FakE news detection method (SAFE) which investigates multi-modal (textual and visual) information of news articles. First, neural networks are adopted to separately extract textual and visual features for news representation. We further investigate the relationship between the extracted features across modalities. Such representations of news textual and visual information along with their relationship are jointly learned and used to predict fake news. The proposed method facilitates recognizing the falsity of news articles based on their text, images, or their "mismatches." We conduct extensive experiments on large-scale real-world data, which demonstrate the effectiveness of the proposed method.

Fake News Early Detection

“…III. Latent features represent news articles via automatically generated features often obtained by matrix/tensor factorization or deep learning techniques, e.g., Text-CNN [25,58,64]. Though these latent features can perform well in detecting fake news, they are often difficult to comprehend, which brings challenges to promote the public's understanding of fake news.…”

“…Content-based fake news detection aims to detect fake news by analyzing the content of news articles. Within a machine learning framework, researchers often detect fake news relying on either latent (via neural networks) [58,64] or non-latent (usually hand-crafted) features [12,40,48,53] of the content (see Section 2 for details). Nevertheless, in all such techniques, fundamental theories in social and forensic psychology have not played a significant role.…”

ReCOVery

“…RST (Rhetorical Structure Theory) organizes a piece of content as a tree that captures the rhetorical relation among its phrases and sentences. We use a pretrained RST parser [8] 23 to obtain the tree for each news article and count each rhetorical relation (in total, 45) within a tree, based on which 45 features are extracted and classified in a traditional statistical learning framework.…”

“…Text-CNN relies on a convolutional neural networks for text classification, which contains a convolutional layer and max pooling. SAFE [23]. 24 SAFE is a neural-network-based method that utilizes news multimodal information for fake news detection, where news representation is learned jointly by news textual and visual 22 https://liwc.wpengine.com/ 23 https://github.com/jiyfeng/DPLP 24 https://github.com/Jindi0/SAFE information along with their relationship.…”

“…The general statistics on our dataset is presented in Table 1. The dataset contains 2,029 news articles, most of which have both textual and visual information for multimodal studies (#=2,017), [23] and have been shared on social media (#=1,747). The dataset is imbalanced in news class -the proportion of reliable versus unreliable news articles is around 2:1.…”