Leveraging Deep Learning and Natural Language Processing for the Identification of Deceptive Online Content

Kamran Dahri; Faheem Ahmed; Muhammad Aquib; Mohib Ali Khan; Muhammad Yaqoob Koondhar

Authors

Kamran Dahri Department of Information Technology, University of Sindh, Jamshoro, Sindh, Pakistan.
Faheem Ahmed Department of Information Technology, University of Sindh, Jamshoro, Sindh, Pakistan.
Muhammad Aquib Department of Information Technology, University of Sindh, Jamshoro, Sindh, Pakistan.
Mohib Ali Khan Department of Information Technology, University of Sindh, Jamshoro, Sindh, Pakistan.
Muhammad Yaqoob Koondhar Information Technology Centre, Sindh Agriculture University.

Keywords:

Fake News Detection, Misinformation, Bidirectional LSTM (Bi-LSTM), Support Vector Machine (SVM), Text Preprocessing, Tokenization, Contextual Embeddings, Deepfake Detection

Abstract

The digital world has a problem with news spreading really fast. Fake news is an issue in the media and news networks. It is a problem that threatens the way we share information around the world. This research paper is about finding a way to automatically detect content. We used machine learning and linguistics, and advanced neural networks to do this. We tried methods to see what works best. We used ways like Logistic Regression and more advanced ways like BERT. We tested our method using the Fake News Net and LIAR datasets. We got good results. Traditional ways, like Support Vector Machines and Random Forests, are still good. Deep learning models work even better. The The BERT model performed well, achieving 92.5% accuracy in detecting fake news. This paper also talks about the problems we face when dealing with news. There are biases in the algorithms. It is hard to understand how the detectors work. We found out that automated detectors are good at what they do. But making them better by improving these models by incorporating contextual understanding remains a challenging problem. If we use what we know about the context, we can help keep conversations safe and make people trust the media again. The digital world has a problem with news spreading really fast. Fake news is an issue in the media and news networks. It is a problem that threatens the way we share information around the world. This research paper is about finding a way to automatically detect content. We used news detection methods and advanced neural networks to do this. We tried fake news detection methods to see what works best. We used ways like Logistic Regression and more advanced ways like BERT to detect fake news. We tested our news detection method using the Fake News Net and LIAR datasets, and we got good results. Traditional machine learning methods such as Support Vector Machines and Random Forests remain competitive at detecting news. Deep learning models work even better at detecting fake news. The BERT model was very good at detecting news stories. It was 92.5 percent of the time when detecting fake news. This paper also talks about the problems we face when dealing with news. There are biases in the news detection algorithms, and it is hard to understand how the fake news detectors work. We found out that automated fake news detectors are good at what they do. But making them better by adding a touch to fake news detection is a hard problem. If we use what we know about the context, we can help keep conversations safe from fake news and make people trust the media again when it comes to fake news.

References

Kai Shu, Amy Sliva, Suhang Wang, Jiliang Tang, Huan Liu, “Fake News Detection on Social Media: A Data Mining Perspective,” arXiv:1708.01967, 2017, [Online]. Available: https://arxiv.org/abs/1708.01967

Y. Yan, P. Zheng, and Y. Wang, “Contrastive Learning-Driven Fake News Detection: Preserving Semantics, Unveiling Distortions,” IEEE Trans. neural networks Learn. Syst., vol. PP, 2025, doi: 10.1109/TNNLS.2025.3634147.

Verónica Pérez-Rosas, Bennett Kleinberg, Alexandra Lefevre, Rada Mihalcea, “Automatic Detection of Fake News,” arXiv:1708.07104, 2017, [Online]. Available: https://arxiv.org/abs/1708.07104

Y. L. Natali Ruchansky, Sungyong Seo, “CSI: A Hybrid Deep Model for Fake News Detection,” arXiv:1703.06959, 2017, [Online]. Available: https://arxiv.org/abs/1703.06959

D. Hosseini and R. Jin, “Graph Neural Network based Approach for Rumor Detection on Social Networks,” 2023 Int. Conf. Smart Appl. Commun. Networking, SmartNets 2023, 2023, doi: 10.1109/SmartNets58706.2023.10215926.

James Thorne, Andreas Vlachos, Christos Christodoulopoulos, Arpit Mittal, “FEVER: a large-scale dataset for Fact Extraction and VERification,” arXiv:1803.05355, 2018, [Online]. Available: https://arxiv.org/abs/1803.05355

X. Zhou and R. Zafarani, “A Survey of Fake News: Fundamental Theories, Detection Methods, and Opportunities,” ACM Comput. Surv., vol. 53, no. 5, Sep. 2020, doi: 10.1145/3395046

Xinyi Zhou, Reza Zafarani, “Network-based Fake News Detection: A Pattern-driven Approach,” arXiv:1906.04210, 2019, [Online]. Available: https://arxiv.org/abs/1906.04210

Yonghyun Jeong, Doyeon Kim, Youngmin Ro, Jongwon Choi, “FrePGAN: Robust Deepfake Detection Using Frequency-level Perturbations,” arXiv:2202.03347, 2022, [Online]. Available: https://arxiv.org/abs/2202.03347

Z. Guo, Z. Jia, L. Wang, D. Wang, G. Yang, and N. Kasabov, “Constructing New Backbone Networks via Space-Frequency Interactive Convolution for Deepfake Detection,” IEEE Trans. Inf. Forensics Secur., vol. 19, pp. 401–413, 2024, doi: 10.1109/TIFS.2023.3324739.

C. Zhao, C. Wang, G. Hu, H. Chen, C. Liu, and J. Tang, “ISTVT: Interpretable Spatial-Temporal Video Transformer for Deepfake Detection,” IEEE Trans. Inf. Forensics Secur., vol. 18, pp. 1335–1348, 2023, doi: 10.1109/TIFS.2023.3239223.

Jumana Jouhar, Anju Pratap, “A Transformer-Based Deep Learning Framework with Semantic Encoding and Syntax-Aware LSTM for Fake Electronic News Detection,” Comput. Mater. Contin., vol. 86, no. 1, pp. 1–25, 2025, doi: https://doi.org/10.32604/cmc.2025.069327.

Zhongjie Ba, Qingyu Liu, Zhenguang Liu, Shuang Wu, Feng Lin, Li Lu, Kui Ren, “Exposing the Deception: Uncovering More Forgery Clues for Deepfake Detection,” arXiv:2403.01786, 2024, [Online]. Available: https://arxiv.org/abs/2403.01786

Davide Alessandro Coccomini, Giorgos Kordopatis Zilos, “MINTIME: Multi-Identity Size-Invariant Video Deepfake Detection,” arXiv:2211.10996, 2022, [Online]. Available: https://arxiv.org/abs/2211.10996

Ruiyang Xia, Decheng Liu, Jie Li, Lin Yuan, Nannan Wang, Xinbo Gao, “MMNet: Multi-Collaboration and Multi-Supervision Network for Sequential Deepfake Detection,” arXiv:2307.02733, 2023, [Online]. Available: https://arxiv.org/abs/2307.02733

N. Vaswani, A., Shazeer, N., Parmar, “Attention is all you need,” Adv. Neural Inf. Process. Syst., vol. 30, pp. 5998–6008, 2017.

C. Peng, Z. Miao, D. Liu, N. Wang, R. Hu, and X. Gao, “Where Deepfakes Gaze at? Spatial-Temporal Gaze Inconsistency Analysis for Video Face Forgery Detection,” IEEE Trans. Inf. Forensics Secur., vol. 19, pp. 4507–4517, 2024, doi: 10.1109/TIFS.2024.3381823.

Y. Yu, X. Liu, R. Ni, S. Yang, Y. Zhao, and A. C. Kot, “PVASS-MDD: Predictive Visual-Audio Alignment Self-Supervision for Multimodal Deepfake Detection,” IEEE Trans. Circuits Syst. Video Technol., vol. 34, no. 8, pp. 6926–6936, 2024, doi: 10.1109/TCSVT.2023.3309899.