Paper Contents
Abstract
Strong prediction and detection systems to protect against cyber hacking breaches are of utmost importance in a period where cybersecurity threats have grown increasingly complex. In order to anticipate and discover potential cyber hacking breaches, this mission provides a unique approach by utilizing Machine Learning techniques, specifically the Random Forest Classifier. The proposed Python device uses a dataset of 5457 URLs with 87 extracted features that has been carefully selected. An important aspect of this dataset is its balanced composition, which is 50% scamming and 50% legal URLs. Correctly identifying cyber risks while limiting false positives is the project's top priority. Executed results demonstrate the system's outstanding performance, which was achieved through thorough training and assessment. With a remarkable 99% training accuracy, the Random Forest Classifier is sure to be able to spot patterns and differentiate between safe and dangerous URLs. Validating that it's trustworthy in real-world circumstances, the edition additionally exhibits a robust test accuracy of 91%. Ultimately, this challenge is a first in the field of internet security breach prediction and detection, using Random Forest Classifier and Machine Learning to provide better protection. Agencies are able to strengthen their cybersecurity defenses against capable cyber threats and attacks, thanks to the significant precision that serves as evidence of its efficiency.
Copyright
Copyright © 2025 Bhoomika C. This is an open access article distributed under the Creative Commons Attribution License.
