A Deep Learning Approach to Fake News Detection

In the extensive growth of social media platforms, fake news is false information spread among people around the world. Fake news weakens public trust and societal stability; it has negative impacts that necessitate for fake news. In order to enhance the public trust and improve the social media users, fake news detection is crucial, which is a complex and multi-dimensional task due to the diverse characteristics of fake news. More researchers have explored several deep learning and machine learning approaches for fake news detection approaches yet the existing binary classification mechanism is not sufficient for handling imbalanced training data to enhance poor performance in the fake news detection process also it affects the interpretability and consumes high computational duration. With the aim of rectifying these complexities, an automated fake news detection mechanism is proposed in this research work to ultimately recognise unreliable information for improving social media users’ effectiveness. Initially, the necessary data are taken from the publicly available datasets, such as the WELFake dataset and Fake News dataset; it is fed into the preprocessing phase. Here, the punctuation, special characters, redundant content and inappropriate data are mitigated, which can generate cleaned data. Then, the preprocessed data is given to the feature extraction process to efficiently extract the relevant informative features for enhancing better detection outcomes. Here, the Bidirectional Encoder Representations from Transformers (BERT) is initially used to retrieve the first set of features, the Recurrent Neural Network (RNN) extracts the temporal features and the Convolution Neural Network (CNN) is used to extract the spatial features. Subsequently, these features are fused in the weighted feature selection process, where the weights are quickly tuned through the Improved Horse Herd Optimisation Algorithm (IHHOA). Further, it is passed to the Residual Convolution Bidirectional Long Short Term Memory (RConv-Bi-LSTM) technique for significantly detecting fake news in a limited duration. The entire detection process is estimated with several performance metrics and contrasted over various conventional methods to generate optimal outcomes. Here, the designed framework’s FM value (dataset 1) is enhanced over 22.9% DNN, 8.7% CNN, 17.6% LSTM and 7% Bi-LSTM methods (BERT features) to show its better performance over the existing approaches.

Recently, fake news has been widely spread through the Internet due to the increased use of social media for communication. Fake news has become a significant concern due to its harmful impact on individual attitudes and the community’s behavior. Researchers and social media service providers have commonly utilized artificial intelligence techniques in the recent few years to rein in fake news propagation. However, fake news detection is challenging due to the use of political language and the high linguistic similarities between real and fake news. In addition, most news sentences are short, therefore finding valuable representative features that machine learning classifiers can use to distinguish between fake and authentic news is difficult because both false and legitimate news have comparable language traits. Existing fake news solutions suffer from low detection performance due to improper representation and model design. This study aims at improving the detection accuracy by proposing a deep ensemble fake news detection model using the sequential deep learning technique. The proposed model was constructed in three phases. In the first phase, features were extracted from news contents, preprocessed using natural language processing techniques, enriched using n-gram, and represented using the term frequency–inverse term frequency technique. In the second phase, an ensemble model based on deep learning was constructed as follows. Multiple binary classifiers were trained using sequential deep learning networks to extract the representative hidden features that could accurately classify news types. In the third phase, a multi-class classifier was constructed based on multilayer perceptron (MLP) and trained using the features extracted from the aggregated outputs of the deep learning-based binary classifiers for final classification. The two popular and well-known datasets (LIAR and ISOT) were used with different classifiers to benchmark the proposed model. Compared with the state-of-the-art models, which use deep contextualized representation with convolutional neural network (CNN), the proposed model shows significant improvements (2.41%) in the overall performance in terms of the F1score for the LIAR dataset, which is more challenging than other datasets. Meanwhile, the proposed model achieves 100% accuracy with ISOT. The study demonstrates that traditional features extracted from news content with proper model design outperform the existing models that were constructed based on text embedding techniques.

Filtering, vetting, and verifying digital information is an area of core interest in information science. Online fake news is a specific type of digital misinformation that poses serious threats to democratic institutions, misguides the public, and can lead to radicalization and violence. Hence, fake news detection is an important problem for information science research. While there have been multiple attempts to identify fake news, most of such efforts have focused on a single modality (e.g., only text‐based or only visual features). However, news articles are increasingly framed as multimodal news stories, and hence, in this work, we propose a multimodal approach combining text and visual analysis of online news stories to automatically detect fake news. Drawing on key theories of information processing and presentation, we identify multiple text and visual features that are associated with fake or credible news articles. We then perform a predictive analysis to detect features most strongly associated with fake news. Next, we combine these features in predictive models using multiple machine‐learning techniques. The experimental results indicate that a multimodal approach outperforms single‐modality approaches, allowing for better fake news detection.

Nowadays, the dissemination of news information has become more rapid, liberal, and open to the public. People can find what they want to know more and more easily from a variety of sources, including traditional news outlets and new social media platforms. However, at a time when our lives are glutted with all kinds of news, we cannot help but doubt the veracity and legitimacy of these news sources; meanwhile, we also need to guard against the possible impact of various forms of fake news. To combat the spread of misinformation, more and more researchers have turned to natural language processing (NLP) approaches for effective fake news detection. However, in the face of increasingly serious fake news events, existing detection methods still need to be continuously improved. This study proposes a modified proof-of-concept model named NER-SA, which integrates natural language processing (NLP) and named entity recognition (NER) to conduct the in-domain and cross-domain analysis of fake news detection with the existing three datasets simultaneously. The named entities associated with any particular news event exist in a finite and available evidence pool. Therefore, entities must be mentioned and recognized in this entity bank in any authentic news articles. A piece of fake news inevitably includes only some entitlements in the entity bank. The false information is deliberately fabricated with fictitious, imaginary, and even unreasonable sentences and content. As a result, there must be differences in statements, writing logic, and style between legitimate news and fake news, meaning that it is possible to successfully detect fake news. We developed a mathematical model and used the simulated annealing algorithm to find the optimal legitimate area. Comparing the detection performance of the NER-SA model with current state-of-the-art models proposed in other studies, we found that the NER-SA model indeed has superior performance in detecting fake news. For in-domain analysis, the accuracy increased by an average of 8.94% on the LIAR dataset and 19.36% on the fake or real news dataset, while the F1-score increased by an average of 24.04% on the LIAR dataset and 19.36% on the fake or real news dataset. In cross-domain analysis, the accuracy and F1-score for the NER-SA model increased by an average of 28.51% and 24.54%, respectively, across six domains in the FakeNews AMT dataset. The findings and implications of this study are further discussed with regard to their significance for improving accuracy, understanding context, and addressing adversarial attacks. The development of stylometric detection based on NLP approaches using NER techniques can improve the effectiveness and applicability of fake news detection.

Detecting fake news on social media is an increasingly important problem, because of the rapid dissemination and detrimental impact of fake news. Graph-based methods that encode news propagation paths into tree structures have been shown to be effective. Existing studies based on such methods represent the propagation of news through static graphs or coarse-grained graph snapshots. They do not capture the full dynamics of graph evolution and hence the temporal news propagation patterns. To address this issue and model dynamic news propagation at a finer-grained level, we propose a temporal graph-based model. We join this model with a neural Hawkes process model to exploit the distinctive self-exciting patterns of true news and fake news on social media. This creates a highly effective fake news detection model that we named SEAGEN. Experimental results on real datasets show that SEAGEN achieves an accuracy of fake news detection of over 93% with an advantage of over 2.5% compared to other state-of-the-art models.

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).

Fake news is a major threat to global democracy resulting in diminished trust in government, journalism and civil society. The public popularity of social media and social networks has caused a contagion of fake news where conspiracy theories, disinformation and extreme views flourish. Detection and mitigation of fake news is one of the fundamental problems of our times and has attracted widespread attention. While fact checking websites such as snopes, politifact and major companies such as Google, Facebook, and Twitter have taken preliminary steps towards addressing fake news, much more remains to be done. As an interdisciplinary topic, various facets of fake news have been studied by communities as diverse as machine learning, databases, journalism, political science and many more. The objective of this tutorial is two-fold. First, we wish to familiarize the database community with the efforts by other communities on combating fake news. We provide a panoramic view of the state-of-the-art of research on various aspects including detection, propagation, mitigation, and intervention of fake news. Next, we provide a concise and intuitive summary of prior research by the database community and discuss how it could be used to counteract fake news. The tutorial covers research from areas such as data integration, truth discovery and fusion, probabilistic databases, knowledge graphs and crowdsourcing from the lens of fake news. Effective tools for addressing fake news could only be built by leveraging the synergistic relationship between database and other research communities. We hope that our tutorial provides an impetus towards such synthesis of ideas and the creation of new ones.

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. .

SummaryIn the present epoch of computing, the world has changed from older conventional print media to social platform channels. Fake news articles have the prospects to handle the opinions of the public and so may harm human groupings. Therefore, it is necessary to explore the authenticity and credibility of the news flash being shared on the internet community. Hence, this research paper devises an efficient and robust fake news detection model, named Exponential Chimp Optimization Algorithm (EChOA)‐based Deep Neuro‐Fuzzy Network (DNFN) for detecting fake news. The introduced model utilizes a MapReduce framework that includes the mapper and reducer phases for processing big data for detecting fake news. First phase of processing is the Mapper work, in which every input used in the database is processed and creates an intermediate key‐value pair. In the reducer phase, the fusion of features is performed by arranging the features with the help of computing the optimal parameter and Rand similarity coefficient using a Deep Q Network (DQN). Here, the detection of fake news is obtained by DNFN, and the DNFN is done using implemented EChOA. The EChOA‐based DNFN effectively generates robust and effective fake news detection performance by choosing the optimal feature subsets through feature fusion. The EChOA is designed by integrating the Exponential Weighted Moving Average (EWMA) and Chimp Optimization Algorithm (ChOA). Moreover, the EChOA‐based DNFN method outperformed various former fake news detection approaches and attains the highest performance based on the testing accuracy is 0.909, sensitivity is 0.937, and specificity is 0.891 using the FakeNewsNet dataset.

The widespread propagation of fake information via social media platforms has been a source of severe concern for misinformation and its potential impact on society. This study compares the Albert-base-v2 transformer and CNN-BiLSTM models to identify fake news on the Fake News Sample-Pontes dataset. The proposed models were trained and evaluated using the Fake News Sample (Pontes) dataset from Kaggle, which includes over 45,000 news articles labeled as real or fake, based on predefined criteria. Preprocessing is done on the dataset by eliminating punctuation, removing non-English characters, and tokenization for improvement in the model's performance. Five deep learning architectures—2-CNN 2-BiLSTM, 3-CNN 1-BiLSTM, 1-CNN 3-BiLSTM, DistilBERT, and Albert-base-v2—are evaluated. The models are trained using a 75%-20%-5% data split, where an embedding size of 300 is used in CNN-BiLSTM architectures. Performance is assessed based on accuracy, precision, recall, F1-score, and AUC-ROC metrics. Among the models, Albert-base-v2 has the best performance with 90.8% accuracy and 0.908 F1-score that outperforms 2-CNN 2-BiLSTM (accuracy of 86.1%, F1-score of 0.861) and DistilBERT (85.0% accuracy, 0.850 F1-score). Statistical significance is determined using t-tests, and class-wise performance is analyzed using a confusion matrix. The results highlight the superiority of transformer-based models over conventional deep learning methods in fake news detection. In addition, limitations, ethical considerations, and future directions toward enhancing model interpretability and efficiency are discussed.

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

Fake news have attracted attention of general public because of the influence they can exert on important activities of society, such as elections. Efforts have been made to detect them, but usually they rely on human labour fact-checking, what can be costly and time consuming. Computational approaches have typically relied on supervised learning models, in which a model is trained based on fake and true news samples. Such approach allows a large amount of news to be classified in a short time, but it demands datasets labelled with positive and negative instances. Our work proposes to detect fake news by training a model with only fake samples in the training dataset, through One-Class Classification (OCC). We compare a novel algorithm, called DCDistanceOCC, to others published in literature, and got similar, or even better, results. The case study is the Brazilian politics scenario starting at the 2018 general elections on Twitter and WhatsApp. These two platforms were a fertile ground to fake news proliferation. We also evaluated the models over another available dataset from literature. To the best of our knowledge, this is the first paper to identify fake news using an OCC approach and also the first one to provide Portuguese-based WhatsApp and Twitter datasets with fake news.

In perceptual decision-making, it is well-known that low confidence in a decision promotes information seeking, which increases subsequent decision accuracy. It is unclear, however, whether these mechanisms generalize to more complex domains such as fake news detection. Here, we tested whether confidence also drives information seeking in the context of fake news, whether seeking information improves fake news detection and sharing intentions, and whether we find confidence-driven confirmation biases. In an online study, 314 US citizens indicated whether news headlines were true or fake and could request additional information before finalizing their decision. First, in line with findings from perceptual decision-making, confidence was the main driver of information seeking for both fake and true news. Yet critically, whereas confidence lawfully scaled with decision accuracy for true news items, we found a U-shaped relationship for fake news items: decision accuracy was lowest for intermediate confidence, indicating that for fake headlines confidence is not a reliable signal of headline veracity. Second, while information seeking increased decision accuracy for true news items, for fake news items decision accuracy remained at chance level even after information seeking. Sharing intentions for fake news remained equally unaffected by information seeking, and mostly increased for belief-congruent true items. Finally, we found a confidence-driven confirmation bias for beliefs, but not for sharing intentions. Together, our results clarify that while confidence-driven information seeking adaptively increases decision accuracy for true news items, it is entirely ineffective in the context of fake news.

Staying up-to-date on news through social media has both advantages and disadvantages. Social media platforms are excellent sources of information, but they also contribute significantly to the spread of fake news (incorrect or poor-quality content). The widespread circulation of such information can profoundly impact society by shaping public opinion, influencing policy debates, and affecting decision-making processes. As a result, fake news detection has become a critical area of research, particularly in multilingual and multicultural contexts. Detecting fake news in Arabic presents unique challenges due to the language’s complexity, diverse dialects, and the limited availability of comprehensive annotated datasets. Many existing detection models are vulnerable to misleading or context-heavy information because they rely primarily on basic content features. This study evaluates the effectiveness of various machine learning and deep learning techniques—including random forest (RF), naive Bayes (NB), feedforward neural networks (FNN), long short-term memory (LSTM), Arabic bidirectional encoder representations from Transformers (BERT), and a hybrid AraBERT + bidirectional long short-term memory (BiLSTM) model—in identifying fake news in Arabic media. Experiments were conducted using the Arabic Fake News Dataset (AFND), which contains over 606,000 news items labeled according to credibility. The hybrid AraBERT + Bi-LSTM model outperformed all others, achieving an average F1-score of 98.0%. This model integrates transformer-based contextual embeddings with sequential modeling.

With the rise of social media, the spread of fake news has become a significant concern, potentially misleading public perceptions and impacting social stability. Although deep learning methods like CNNs, RNNs, and Transformer-based models like BERT have enhanced fake news detection. However, they primarily focus on content and do not consider social context during news propagation. Graph-based techniques have incorporated the social context but are limited by the need for large labeled datasets. To address these challenges, this paper introduces GAMC, an unsupervised fake news detection technique using the Graph Autoencoder with Masking and Contrastive learning. By leveraging both the context and content of news propagation as self-supervised signals, our method reduces the dependency on labeled datasets. Specifically, GAMC begins by applying data augmentation to the original news propagation graphs. Subsequently, these augmented graphs are encoded using a graph encoder and subsequently reconstructed via a graph decoder. Finally, a composite loss function that encompasses both reconstruction error and contrastive loss is designed. Firstly, it ensures the model can effectively capture the latent features, based on minimizing the discrepancy between reconstructed and original graph representations. Secondly, it aligns the representations of augmented graphs that originate from the same source. Experiments on the real-world dataset validate the effectiveness of our method.