Osirim Love Eseroghene
Research Subjects
Research Interests
Osirim Love Eseroghene has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
- Abiodun Theresa Nkechi
Department of Computer and Information Sciences, Covenant University, Ota
1 shared publication - Jegede Tope John
Department of Computer and Information Sciences, Covenant University, Ota
1 shared publication - Odunayo Damilola Osofuye
Department of Computer and Information Sciences, Covenant University
1 shared publication
Latest Publications
- Open Access
- 7 January 2026
- Published by EDP Sciences in E3S Web of Conferences
- Vol. 684, 03001
Development of Malware Detection Web Plugin Using One-Dimensional Convolutional Neural Networks Model
The internet has become increasingly essential in daily life, yet threats to computer systems, networks, and personal privacy continue to rise. Many users turn to unverified or illegal websites to bypass access restrictions, exposing themselves to embedded malware or malicious downloadable files. This study addresses this challenge by developing a web plugin capable of detecting malware on websites and files in real time. The plugin integrates a One-Dimensional Convolutional Neural Network (1D-CNN) model to analyze sequential patterns extracted from webpage structures and file headers. The 1D-CNN model demonstrated strong performance across both datasets, achieving 0.88 accuracy, 0.88 precision, 0.89 recall, and 0.88 F1-score for web-based threat detection, and 0.98 accuracy, 0.99 precision, 0.98 recall, and 0.98 F1-score for file-based malware detection. These results highlight the model's reliability for realtime website and file scanning, empowering users to make safer and more informed browsing decisions.
- 7 January 2026
- Published by EDP Sciences in E3S Web of Conferences
- Vol. 684, 03001
Open AccessDevelopment of Malware Detection Web Plugin Using One-Dimensional Convolutional Neural Networks Model
The internet has become increasingly essential in daily life, yet threats to computer systems, networks, and personal privacy continue to rise. Many users turn to unverified or illegal websites to bypass access restrictions, exposing themselves to embedded malware or malicious downloadable files. This study addresses this challenge by developing a web plugin capable of detecting malware on websites and files in real time. The plugin integrates a One-Dimensional Convolutional Neural Network (1D-CNN) model to analyze sequential patterns extracted from webpage structures and file headers. The 1D-CNN model demonstrated strong performance across both datasets, achieving 0.88 accuracy, 0.88 precision, 0.89 recall, and 0.88 F1-score for web-based threat detection, and 0.98 accuracy, 0.99 precision, 0.98 recall, and 0.98 F1-score for file-based malware detection. These results highlight the model's reliability for realtime website and file scanning, empowering users to make safer and more informed browsing decisions.