A High Accuracy Classifier-Based Approach for Detecting Phishing on Social Media

Olaiya; Olayinka Oluwaseun; Aderanti Aderemi Taiwo; Moshood; Babatunde Ismail

doi:10.56201/ijemt.vol.12.no1.2026.pg344.354

📄 Download Paper

A High Accuracy Classifier-Based Approach for Detecting Phishing on Social Media

Olaiya, Olayinka Oluwaseun, Aderanti Aderemi Taiwo, Moshood, Babatunde Ismail

Abstract

The increasing prevalence of phishing attacks on social media platforms poses a serious challenge to online security and user trust. Cybercriminals exploit the openness and anonymity of these platforms to deceive users into revealing sensitive information or downloading malicious content. This study presents a high-accuracy classifier-based approach for detecting phishing on social media by leveraging machine learning and natural language processing techniques. The system collects and preprocesses textual and URL-based data from social platforms, extracting both lexical and behavioral features to train multiple classifiers such as Random Forest, Support Vector Machine , and Gradient Boosting. A comparative evaluation of these algorithms was conducted using a synthetic dataset and real-world data samples. The results show that ensemble-based models, particularly the Random Forest and Gradient Boosting classifiers, achieved superior detection accuracy, precision, and recall compared to individual learners. The integration of NLP-driven feature extraction improved the model’s ability to identify deceptive linguistic cues and malicious intent embedded within phishing messages. The developed system was implemented using Python and Streamlit for real- time analysis, offering an interactive interface for detecting and classifying suspicious posts and URLs on social media platforms. The findings highlight that machine learning-based approaches, when combined with linguistic analysis and ensemble learning, can significantly enhance phishing detection accuracy and reduce false positives. This research contributes to the growing body of cybersecurity solutions by demonstrating a practical, scalable, and intelligent framework for early phishing detection in dynamic online environments.

Keywords

CybersecurityEnsemble LearningMachine LearningPhishing DetectionSocial Media

References

Acar, G., Mathur, A., Huang, L., & Narayanan, A. (2024). Phishing in the age of social media: Human factors and detection challenges. Computers & Security, 137, 103972. Alghamdi, T., Khan, R. A., & Shaikh, A. (2024). Generative AI and its impact on phishing detection and prevention. IEEE Access, 12, 77712–77726. Alsubaei, F., Al-Saleh, M., & Khan, S. (2021). Intelligent URL-based phishing detection using hybrid machine learning. Computers & Security, 110, 102404. Alzubaidi, M., Al-Qudah, M., & Al-Rousan, N. (2022). Detecting phishing attacks using deep learning and natural language processing. Journal of Network and Computer Applications, 202, 103371. Banu, S., & Usha, R. (2021). Hybrid feature-based phishing detection using machine learning. ICT Express, 7(4), 512–519. Borate, V., Adsul, A., Dhakane, R., Gawade, S., Ghodake, S., & Jadhav, P. (2023). A comprehensive review of phishing attack detection using machine-learning techniques. Journal of Discrete Mathematical Sciences & Cryptography, 25(3), 817–827. Borgohain, R., & Das, S. (2024). Phish-Guard: Ensemble deep learning framework for phishing detection on social platforms. Expert Systems with Applications, 235, 121000. Hammed A.O & Soyemi, J.O. (2022). Defending against phishing attacks in social networks using AI-based approaches. Future Generation Computer Systems, 129, 219–231. Huang, X., Cao, Y., & Zhang, Q. (2024). Sharing card forgery: A new phishing vector across social-media platforms. IEEE Transactions on Information Forensics and Security, 19, 2041–2055. Patil, A., & Pawar, R. (2022). Detection of phishing websites using machine learning: A comparative study. International Journal of Information Technology, 14, 567–579. Safi, A., Ahmad, M., & Li, J. (2023). A classifier-based ensemble model for phishing URL detection using natural language features. Computers and Electrical Engineering, 106, 108636. Wilk-Jakubowski, J. L., Pawlik, L., Wilk-Jakubowski, G., & Sikora, A. (2025). Machine learning and neural networks for phishing detection: A systematic review (2017–2024). Electronics, 14(18), 3744. Yang, L., Hu, P., & Cui, H. (2024). Adaptive learning framework for phishing detection on social media using multimodal data. Neural Networks, 176, 106157. Zhang, H., Wang, L., & Li, R. (2023). Deploying phishing detection systems with Streamlit and Python-based machine learning. SoftwareX, 22, 101322.

A High Accuracy Classifier-Based Approach for Detecting Phishing on Social Media

Abstract

Keywords

References

More Articles from INTERNATIONAL JOURNAL OF ENGINEERING AND MODERN TECHNOLOGY