Aug 2026· Journal of Intelligent Decision Making and Information Science· 0 citations· 19 references
TL;DR
A Four-Level Hierarchical Attention Network (4HAN) that incorporates word-, sentence-, and headline-level attention, along with Hypergraph Convolution and Hypergraph Attention, is proposed using the LIAR dataset and shows a detection accuracy rate of 96.00%, which beats multiple existing methodologies in fake news detection.
Abstract
Due to the fast proliferation of online news media and social networks, there is a significant increase in the dissemination of misleading data and fake news on digital platforms. Fake news detection is difficult due to the incapacity of deep learning models or conventional machine learning in comprehending semantic and higher-level interactions between news text, news author, publisher, and additional metadata. This results in decreased detection efficiency and enables the dissemination of misinformation that impacts societal behavior and public opinion. To address this problem, a Four-Level Hierarchical Attention Network (4HAN) that incorporates word-, sentence-, and headline-level attention, along with Hypergraph Convolution and Hypergraph Attention, is proposed using the LIAR dataset. By combining semantic feature extraction and relational dependency modeling, the 4HAN framework achieves better results. Results showed a detection accuracy rate of 96.00%, which beats multiple existing methodologies in fake news detection. These findings demonstrate that combining hierarchical attention with hypergraph learning provides more effective semantic and relational representation, leading to improved fake news detection performance and greater robustness in misinformation analysis.
A multi-model learning framework that combines the complementary strengths of classical machine learning classifiers, deep sequential neural networks, and transformer-based contextual language models to detect fake news on social media is proposed.
Priya Verma· International Journal of Res...· 0 citations
Heterogeneous Graph Attention Network (HGAT) is proposed, where a pretrained BERT-Large encoder is coupled with a Heterogeneous Graph Attention Network (HGAT) to learn joint representations for textual, social network and external knowledge graph features.
Akash Garg, Sachin Pachauri· Journal of Artificial Intell...· 0 citations
The digital news portals and social media are rapidly expanding, which has significantly increased the spread of fake news, which affects public opinion, social harmony, and trust in information sources. Detection of fake news at an early stage is a critical research challenge. In recent years, researchers have applied multiple methods for the classification of news articles as fake using Natural Language Processing (NLP), Machine Learning (ML), and Deep Learning (DL) techniques. This paper discusses a feature-based analysis of text-oriented fake news detection methods published from 2017 to 2025. The analysis demonstrates how different textual features, such as linguistic, stylistic, psychological, statistical, semantic, and syntactic features, are used for the identification of fake or real news. A comparative analysis of existing studies shows that most research primarily depends on statistical and semantic representations like N-grams, TF-IDF, and word embeddings, whereas linguistic, stylistic, and psychological cues are comparatively less explored. In addition, syntactic features have gained very limited attention despite their potential to enhance detection performance. The review emphasizes integrating multiple feature types to develop more reliable and interpretable detection systems. It also identifies research gaps and suggests future directions for developing comprehensive feature-based frameworks for fake news detection
Unknown authors· International Research Journ...· 0 citations
The increasing prevalence of fake news on social media has raised serious concerns due to its impact on public perception and decision-making. In response, this study introduces a hybrid ensemble learning framework enhanced by RoBERTa-based feature representation and optimized using the Whale Optimization Algorithm (WOA). The proposed approach aims to effectively capture deep semantic patterns in textual data while improving classification performance through adaptive parameter tuning. RoBERTa is employed to generate high-quality textual embeddings from news content, which are then fed into a set of base classifiers to ensure robustness and diversity in predictions. The WOA algorithm fine-tunes the ensemble model parameters, resulting in improved convergence and reduced error rates. The model was evaluated using two well-known fake news datasets, LIAR and ISOT. On the LIAR dataset, the proposed method achieved an accuracy of 98.17%, precision of 98.1%, recall of 97.8%, and an F1-score of 97.9%. On the ISOT dataset, it achieved an accuracy of 99.24%, precision of 99.3%, recall of 98.9%, and an F1-score of 99.1%. These results confirm the high reliability and balanced performance of the framework in distinguishing between true and false information across diverse content.
Kareem Awad Dawood, Golnaz Aghaee Ghazvini, Ali Albu-Rghaif et al.· Digital Transformation and A...· 0 citations
Comparative analysis demonstrates that the combination of deep learning and knowledge of the environment significantly enhances the level of detection in domains.
Pundlik Dattatray Jadhav, R. K. Shukla· international journal of eng...· 0 citations
This paper proposes modeling data through ego networks, extracting features from the underlying network, matching with textual features, and using traditional machine learning algorithms to detect and identify fake news on social networks.
Jhonathan Carvalho, B. Travençolo, F. Pereira· Anais do XV Brazilian Worksh...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
