A Review of Previous Research on Machine  Learning, Deep Learning, and Natural  Language Processing for Text-Based  Fake News Detection

Akram Saeed Aqlan Alhammadi; Dalal Q.A Saif; Al-jaberi Ramzi Hamid

A Review of Previous Research on Machine Learning, Deep Learning, and Natural Language Processing for Text-Based Fake News Detection

المؤلفون

Akram Saeed Aqlan Alhammadi Information Technology department, College of Information Technology and Computer Science University of Saba Region, Marib, Yemen
Dalal Q.A Saif Faculty of Engineering and Information Technology, Taiz University, Taiz, Yemen
Al-jaberi Ramzi Hamid Information System department University of Saba Region, Marib, Yemen

الكلمات المفتاحية:

Fake News Detection، Social Media، Deep Learning، chine Learning، NLP

الملخص

The spread of misinformation through text has become a critical problem in the current digital age, impacting public opinion and the credibility of the media. To maintain the integrity of information contained in texts, it is essential to identify and counter misinformation. This study examines twelve previous studies that employed machine learning, deep learning, and natural language processing methods to detect fake news in texts and were published between 2020 and 2024. The study aims to highlight the key techniques for identifying false news, assess their effectiveness, and pinpoint research gaps. The results demonstrated that whereas classic machine learning algorithms like SVM achieved an accuracy of 95.05%, deep learning models like BERT and BiLSTM produced higher accuracy, reaching up to 98.90%. The study also identified key challenges including a lack of standardized benchmarks, generalization issues, and data bias. In order to close these gaps and raise the precision and effectiveness of detection systems, the paper concludes by proposing directions for future research.