A review of deep-learning-based models for afaan oromo fake news detection on social media networks

Abstract

Abstract This study explores fake news detection techniques in Afan Oromo using deep learning methods. It compares machine learning and deep learning algorithms, proposes future studies using picture data, and encourages more effective methodologies. The review aims to assess current deep learning models for fake news detection in Afaan Oromo, identify challenges, compare techniques, review available datasets, analyze performance metrics, provide recommendations for future research, discuss societal implications, explore integration with other technologies, address ethical considerations, and provide cross-language insights. It also explores potential integrations and ethical considerations. This review examines the methods employed in detecting fake news in Afaan Oromo on social media platforms. Key components include a literature review, dataset compilation, preprocessing, feature extraction, model selection, training and validation, evaluation metrics, results analysis, discussion, and future work recommendations. The study aims to provide a comprehensive understanding of how deep-learning-based models can detect fake news in Afaan Oromo and address unique challenges in low-resource settings. The paper analyzes various deep-learning models for detecting fake news in the Afaan Oromo language. It discusses model performance, data challenges, feature extraction techniques, implications for social media, and future directions, including the need for robust datasets, improved model architectures, and cross-linguistic studies. The study examines fake news detection, comparing traditional machine learning methods with modern deep learning techniques. It focuses on three categories of deceptive information: liars, fake news publications, and corpus datasets. The findings suggest the need for specialized datasets and advanced models to mitigate fake news spread and enhance information credibility.