A Deep Learning-Based Approach for Detecting Fraudulent Websites in Modern Cybersecurity Systems
INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT · 11.04.2026 · 10 (04) · 1-9
DOI: 10.55041/ijsrem59873 ↗Kurz gefasst
Fachartikel in der Zeitschrift INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT, online erschienen am 11.04.2026. Verfasst von Daamodara Durga Prasad und 4 weiteren Autor:innen.
Abstract (Originalsprache)
Abstract—Fraudulent websites pose a significant threat to modern cybersecurity systems, targeting users through phishing attacks, credential theft, and financial fraud. Traditional rule- based detection methods struggle to keep pace with the evolving sophistication of malicious actors. This paper presents a compre- hensive analysis of deep learning-based approaches for detect- ing fraudulent websites, examining state-of-the-art architectures including Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM) networks, hybrid models, and transformer-based systems. We systematically review methodologies that leverage URL features, HTML content analysis, visual similarity detection, and semantic embeddings to identify malicious websites. Our analysis reveals that deep learning models achieve detection accuracies ranging from 93% to 99.9%, with transformer-augmented architectures and optimized 1D-CNNs reporting the highest performance. We examine common datasets including PhishTank, UCI repository, and Mendeley collections, and discuss the impact of feature en- gineering, ensemble methods, and hyperparameter optimization on model performance. The findings demonstrate that hybrid approaches combining multiple feature types and architectures offer superior robustness and generalization. This work provides insights into current best practices and identifies future research directions for enhancing fraudulent website detection in dynamic threat landscapes. Keywords: Deep Learning, Fraudulent Website Detection, Phishing Detection, Convolutional Neural Networks, LSTM, Transformer Models, Cybersecurity, URL Analysis, Feature En- gineering, Hybrid Models Index Terms—component, formatting, style, styling, insert
Zitieren
APA
Prasad, D. D., vinay, Y. N., Reddy, K. D., Kumar, K. J. S. V. S. V., & teja, N. S. (2026). A Deep Learning-Based Approach for Detecting Fraudulent Websites in Modern Cybersecurity Systems. INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT, 10(04), 1-9. https://doi.org/10.55041/ijsrem59873
Vancouver
Prasad DD, vinay YN, Reddy KD, Kumar KJSVSV, teja NS. A Deep Learning-Based Approach for Detecting Fraudulent Websites in Modern Cybersecurity Systems. INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT. 2026;10(04):1-9. doi:10.55041/ijsrem59873
BibTeX
@article{Prasad2026,
author = {Prasad, Daamodara Durga and vinay, Yarramsetti Naga and Reddy, Kallam Dinesh and Kumar, Kathinkkula J S V S Vyshnav and teja, Nalamala Surya},
title = {A Deep Learning-Based Approach for Detecting Fraudulent Websites in Modern Cybersecurity Systems},
journal = {INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT},
year = {2026},
volume = {10},
number = {04},
pages = {1--9},
doi = {10.55041/ijsrem59873}
}
EUVIMED