A Hybrid Approach for Fake News Detection on Cloud Computing | ||
| International Journal of Theoretical and Applied Research | ||
| Articles in Press, Corrected Proof, Available Online from 05 September 2025 PDF (1.07 M) | ||
| Document Type: Original Article | ||
| DOI: 10.21608/ijtar.2025.402732.1136 | ||
| Authors | ||
| Abdullah Ahmed Abdullah* 1; Dina R. Salem1; AbdAllah A. AlHabshy2; Kamal A. ElDahshan2 | ||
| 1School of Computer Science, Canadian International College – CIC, Cairo, Egypt | ||
| 2Department of Mathematics, Faculty of Science, Al-Azhar University, Cairo, Egypt | ||
| Abstract | ||
| The rapid spread of fake news on digital platforms has become a global crisis that threatens public trust and decision-making. Some people make it up for attention or political gain. Machine learning and deep learning techniques have been developed to detect fake news. However, they tend to generate inaccurate reports. To detect fake news, this paper proposes a hybrid model that combines CNN and LSTM frameworks on Google Cloud. This model was able to categorize news with better accuracy than using each model individually. The model was tested and trained on a fake news classification dataset. We used different evaluation metrics (precision, recall, F1 metric, etc.) to measure the efficiency of the model. The democratization of content creation through social media platforms, blogs, and online news portals has enabled unprecedented access to real-time information. We believe that our hybrid classifier-based system has a higher level of reliability. The results and discussion section provides evidence for our claims and establishes the objectives. | ||
| Keywords | ||
| Fake news; Machine learning; Deep learning; CNN; LSTM | ||
|
Statistics Article View: 97 PDF Download: 87 |
||
