,
Research Scholar, School of Computer Science Engineering and Information Systems, Vellore of Technology , Vellore, Tamil Nadu , India
Professor, School of Computer Sciences and Engineering, Vellore Institute of Technology , Vellore, Tamil Nadu , India
Cyberbullying are becoming more susceptible to online social networks because of the large-scale user
generated content. The current methods of detection are primarily post-event methods and lack built-in
prevention strategies, thereby limiting their ability to ensure the protection of user privacy and platform
security. In this paper, a Horned Lizard CatBoost Framework (HLCF) is proposed to predict and prevent
cyber threats active in social networks in advance. It uses the Horned Lizard Optimisation of adaptive
feature selection with a CatBoost-based classifier to precisely differentiate between malicious and non
malicious activity. The existing approaches lack methods that combine feature selection with
optimisation, and a preventive mechanism for access-control. The framework was tested with benchmark
social media datasets, which comprised over 47000 instances, and cross-validation and standard train
test splits. The findings indicate that optimised threat prediction gained 99.98% accuracy prediction,
99.98% precision, 99.98% recall, and a 99.98% F1score. Meanwhile, it reduced the error ratio by
0.0001%. . The suggested HLCF provides a scalable solution for improving security and privacy. This
work demonstrates the efficiency of combining bio-inspired optimisation with machine learning for
cyberbullying prevention in social networks.
This is an open access article distributed under the Creative Commons Attribution Non-Commercial License (CC BY-NC) License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
0
The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.