Advanced Text Analytics - Graph Neural Network for Fake News Detection in Social Media

Fake news detection within online social media using supervised artificial intelligence algorithms

Multi-view co-attention network for fake news detection by modeling topic-specific user and news source credibility

As social media becomes more and more popular, fake news spreads rapidly which is more likely to cause serious consequences, especially during the COVID-19 pandemic. On the premise of meeting data privacy and security requirements, federated learning uses multi-party heterogeneous data to further promote machine learning. This paper proposes a federal learning based COVID-19 fake news detection model with deep self-attention network (FL_FNDM). We construct a deep self-attention network for fake news detection, which combines self-attention-based pretrained model BERT and deep convolutional neural network to detect fake news. Moreover, the fake news detection model is learned under the framework of horizontal federated learning, aiming at protecting users’ data security and privacy. The experimental results demonstrate that the proposed model can improve the performance of fake news detection on the COVID-19 dataset, which can achieve almost the same effect of sharing data without leaking user data.

Easy access, variety of content, and fast widespread interactions are some of the reasons making social media increasingly popular. However, this rise has also enabled the propagation of fake news, text published by news sources with an intent to spread misinformation and sway beliefs. Detecting it is an important and challenging problem to prevent large scale misinformation and maintain a healthy society. We view fake news detection as reasoning over the relations between sources, articles they publish, and engaging users on social media in a graph framework. After embedding this information, we formulate inference operators which augment the graph edges by revealing unobserved interactions between its elements, such as similarity between documents’ contents and users’ engagement patterns. Our experiments over two challenging fake news detection tasks show that using inference operators leads to a better understanding of the social media framework enabling fake news spread, resulting in improved performance.

Fake news is a growing problem on social media and can have significant negative consequences for individuals and society owing to its accessibility, low cost, and quick distribution. It is difficult to automatically identify bogus news that defies the current content-based analysis techniques. One of the key reasons is that current NLP algorithms still lack common sense, which is frequently necessary for understanding how to read the news. Recent research has demonstrated that the propagation patterns of true and fake news differ on social media. With the potential for automatic fake news detection, this study investigates the use of Deep Learning (DL) models to detect fake news on social media. A Neural Network (NN) model was developed using Natural Language Toolkit (NLTK), TensorFlow, and Natural Language Processing (NLP) for textual analysis. The model was trained on a dataset of fake and real news articles, and was able to achieve high accuracy in identifying fake news. The results of this study suggest that DL models can be valuable tools for detecting fake news on social media.

Social networks have become the main platforms for information dissemination. Nevertheless, due to the increasing number of users, social media platforms tend to be highly vulnerable to the propagation of disinformation - making the detection of fake news a challenging task. In this work, we focus on content-based methods for detecting fake news - casting the problem to a binary text classification one (an article corresponds to either fake news or not). In particular, our work proposes a graph-based semi-supervised fake news detection method based on graph neural networks. The experimental results indicate that the proposed methodology achieves better performance compared to traditional classification techniques, especially when trained on limited number of labeled articles 11Our code is publicly available at: https://github.com/bdvllrs/misinformation-detection-tensor-embeddings..

The advancement of internet technology has facilitated the spread of information, including false information or fake news. The dissemination of hoaxes on social media, such as Twitter, can cause confusion and negatively impact society. This study aims to implement a hybrid model that combines Bidirectional Long Short-Term Memory (Bi-LSTM) and Convolutional Neural Network (CNN) for hoax detection. The dataset used consists of English tweets containing both real and fake news, collected between 2020 and 2022, as provided by the TruthSeeker dataset. The model utilizes an embedding layer with word2vec, a Conv1D layer, and a BiLSTM layer to effectively capture temporal and spatial patterns in text data. Additionally, experiments were conducted by varying the number of BiLSTM units and CNN filters to analyze their impact on model performance. After conducting parameter experiments, the best results were achieved using a Conv1D layer with 64 filters and a BiLSTM layer with 64 neurons/units. The evaluation results on the test data indicate an accuracy of 96.14%, a precision of 96%, a recall of 96.25%, and an F1-score of 96%. These results demonstrate the model's high capability in accurately detecting hoaxes, which is significant for combating misinformation on social media. With its strong performance, the model has potential applications in real-time content moderation systems, early hoax detection tools, and digital literacy platforms to help reduce the spread of false information.

With the increasing popularity of news reading on social media platforms, the number of fake news continues to increase, which has a serious impact on society. Since fake news is easier to spread on social media, automatic identification of fake news on social media has become an important research field. In this paper, we propose a Topic-aware Graph Neural Network (TGNN) to identify fake news on social media sites. We first leverage a Semantics-assisted Non-negative Matrix Factorization (SeaNMF) topic model to capture the topic knowledge of posts. Secondly, TGNN integrates topic knowledge into a graph neural network to extract textual features of posts, which are used to identify fake news. The experimental results on a real-world dataset show that, compared with the state-of-the-art baseline models, our model achieves the best performance in three evaluation metrics, which verifies the effectiveness of TGNN.

The easy sharing of multimedia content on social media has caused a rapid dissemination of fake news, which threatens society’s stability and security. Therefore, fake news detection has garnered extensive research interest in the field of social forensics. Current methods primarily concentrate on the integration of textual and visual features but fail to effectively exploit multi-modal information at both fine-grained and coarse-grained levels. Furthermore, they suffer from an ambiguity problem due to a lack of correlation between modalities or a contradiction between the decisions made by each modality. To overcome these challenges, we present a Multi-grained Multi-modal Fusion Network (MMFN) for fake news detection. Inspired by the multi-grained process of human assessment of news authenticity, we respectively employ two Transformer-based pre-trained models to encode token-level features from text and images. The multi-modal module fuses fine-grained features, taking into account coarse-grained features encoded by the CLIP encoder. To address the ambiguity problem, we design uni-modal branches with similarity-based weighting to adaptively adjust the use of multi-modal features. Experimental results demonstrate that the proposed framework outperforms state-of-the-art methods on three prevalent datasets.

Social media is a remarkable outcome of Web 2.0 technology, which is very popular among the Internet users. The general public is using social media as a communication media to fulfil their information requirements on various occasions such as disaster communication, health communication, marketing products and services and political campaigns. However, the openness of social media provides a great platform for misinformation sharing which is a very common problem in social media communication. Hence, the main objectives of this study are to analyse the publication year and total citation count of publications on misinformation on social media and to identify the main disciplines of misinformation studies on social media using the text mining technique. The primary data of this study were extracted from the Web of Science database using the keywords; social media, misinformation, disinformation and fake news on 16th April 2020. The WoS provided 62 search results and all 62 articles were considered in this study. The title of the article, journal, published year, total citation, abstract, author keywords, keywords plus, Web of Science categories, and research areas were extracted from WoS database. The text mining was done manually. According to the results, scholarly publications on misinformation on social media were first published in the year 2012. Scholarly publications were categorized into 10 main categories; Information, Media, Medical information, Social Science, Communication, Health information, Computer science, Other Sciences, Engineering and Management and Finance. The medical information subject area is covering vast varieties of research areas than the other main subject areas.

Due to the widespread use of social media, people are exposed to fake news and misinformation. Spreading fake news has adverse effects on both the general public and governments. This issue motivated researchers to utilize advanced natural language processing concepts to detect such misinformation in social media. Despite the recent research studies that only focused on semantic features extracted by deep contextualized text representation models, we aim to show that content-based feature engineering can enhance the semantic models in a complex task like fake news detection. These features can provide valuable information from different aspects of input texts and assist our neural classifier in detecting fake and real news more accurately than using semantic features. To substantiate the effectiveness of feature engineering besides semantic features, we proposed a deep neural architecture in which three parallel convolutional neural network (CNN) layers extract semantic features from contextual representation vectors. Then, semantic and content-based features are fed to a fully connected layer. We evaluated our model on an English dataset about the COVID-19 pandemic and a domain-independent Persian fake news dataset (TAJ). Our experiments on the English COVID-19 dataset show 4.16% and 4.02% improvement in accuracy and f1-score, respectively, compared to the baseline model, which does not benefit from the content-based features. We also achieved 2.01% and 0.69% improvement in accuracy and f1-score, respectively, compared to the state-of-the-art results reported by Shifath et al. (A transformer based approach for fighting covid-19 fake news, arXiv preprint arXiv:2101.12027, 2021). Our model outperformed the baseline on the TAJ dataset by improving accuracy and f1-score metrics by 1.89% and 1.74%, respectively. The model also shows 2.13% and 1.6% improvement in accuracy and f1-score, respectively, compared to the state-of-the-art model proposed by Samadi et al. (ACM Trans Asian Low-Resour Lang Inf Process, https://doi.org/10.1145/3472620, 2021).

PurposeThis study aims to understand the allure and danger of fake news in social media environments and propose a theoretical model of the phenomenon.Design/methodology/approachThis qualitative research study used the uses and gratifications theory (UGT) approach to analyze how and why people used social media during the 2016 US presidential election.FindingsThe thematic analysis revealed people were gratified after using social media to connect with friends and family and to gather and share information and after using it as a vehicle of expression. Participants found a significant number of fake news stories on social media during the 2016 US presidential election. Participants tried to differentiate between fake news and real news using fact-checking websites and news sources and interacted with the social media users who posted fake news and became part of the echo chamber. Behaviors like these emerged in the analysis that could not be completely explained by UGT and required further exploration which resulted in a model that became the core of this study.Research limitations/implicationsThis is a small-scale exploratory study with eight diverse participants, findings should not be generalized to larger populations. Time-specific self-reporting of information from social media and fake news during the 2016 US presidential election. Upgrading public policies related to social media is recommended in the study, contributing to burgeoning policy discussions and provides recommendations for both purveyors of social media and public policymakers.Practical implicationsUpgrade in public policies related to social media is recommended in the study and contributes to burgeoning policy discussions and provides recommendations for both purveyors of social media and public policymakers.Social implicationsSocial media users are spending increased time on their preferred platforms. This study increases the understanding of the nature, function and transformation of virtual social media environments and their effects on real individuals, cultures and societies.What is original/of value about the paper?This exploratory study establishes the foundation on which to expand research in the area of social media use and fake news.Originality/valueThis exploratory study establishes the foundation to expand research in the area of social media use and fake news.

As news reading on social media becomes more and more popular, fake news becomes a major issue concerning the public and government. The fake news can take advantage of multimedia content to mislead readers and get dissemination, which can cause negative effects or even manipulate the public events. One of the unique challenges for fake news detection on social media is how to identify fake news on newly emerged events. Unfortunately, most of the existing approaches can hardly handle this challenge, since they tend to learn event-specific features that can not be transferred to unseen events. In order to address this issue, we propose an end-to-end framework named Event Adversarial Neural Network (EANN), which can derive event-invariant features and thus benefit the detection of fake news on newly arrived events. It consists of three main components: the multi-modal feature extractor, the fake news detector, and the event discriminator. The multi-modal feature extractor is responsible for extracting the textual and visual features from posts. It cooperates with the fake news detector to learn the discriminable representation for the detection of fake news. The role of event discriminator is to remove the event-specific features and keep shared features among events. Extensive experiments are conducted on multimedia datasets collected from Weibo and Twitter. The experimental results show our proposed EANN model can outperform the state-of-the-art methods, and learn transferable feature representations.

Fake news detection on job advertisements has grabbed the attention of many researchers over past decade. Various classifiers such as Support Vector Machine (SVM), XGBoost Classifier and Random Forest (RF) methods are greatly utilized for fake and real news detection pertaining to job advertisement posts in social media. Bi-Directional Long Short-Term Memory (Bi-LSTM) classifier is greatly utilized for learning word representations in lower-dimensional vector space and learning significant words word embedding or terms revealed through Word embedding algorithm. The fake news detection is greatly achieved along with real news on job post from online social media is achieved by Bi-LSTM classifier and thereby evaluating corresponding performance. The performance metrics such as Precision, Recall, F1-score, and Accuracy are assessed for effectiveness by fraudulency based on job posts. The outcome infers the effectiveness and prominence of features for detecting false news. .

In this manuscript, the rise of misinformation and harmful behavior on social media platforms has made it crucial to develop systems that ensure the authenticity of shared content and user accounts is presented. Parsing of social media feeds for forensic investigation aims to identify fake news and detect fake social media accounts (focusing on platforms like Instagram). The main goal of parsing of social media feeds for forensic investigation is to protect users from misleading information spread by fake accounts. To achieve this, various modules are integrated into the system, including Data Collection and Pre-Processing, Fake News Detection and Account Authenticity Verification. Web scraping tools such as BeautifulSoup are used to gather real-time data, while machine learning algorithms, including Random Forest and Large Language Model (LLM) are employed to classify fake news and detect suspicious account behaviors. The system also uses NLP models to identify fake news patterns and generate comprehensive reports summarizing the findings of each module. By combining Natural Language Processing (NLP) and machine learning techniques, parsing of social media feeds for forensic investigation provides an essential platform for users to protect themselves from misleading information on social networks.