The ever-changing threats in the field of cybersecurity calls for creative solutions when it comes to identifying and mitigating malicious URLs. This issue is resolved by machine learning algorithms’ inherent limitation to solely accept numerical inputs, despite raw URLs with nonnumeric characters. This research study employed the XGBoost classifier and achieved an accuracy of 94.33% in performing the classification task. The proposed approach transforms raw URLs into numerical lexical data, enabling the utilization in machine learning methodologies. To enhance the Confidentiality, Integrity, and Availability (CIA) triads of applications and elevate authentication standards, this research study incorporates specialized mechanisms to target malicious URLs. These mechanisms operate concurrently, focusing on detecting and analyzing malicious URL activity within controlled environments. This study intends to interpret intricate attack patterns, entry points, and tactics employed by potential attackers. Integrating machine learning-driven URL classification with the analysis of malicious URL behaviors fosters a comprehensive approach aimed at enhancing cybersecurity defenses. By furnishing insights into emerging threats and bolstering enterprises against evolving cyber risks, this synergy endeavors to fortify cybersecurity postures. A complete strategy to improve cybersecurity defences is fostered by combining the analysis of dangerous URL behaviours with machine learning-driven URL classification. Strengthening cybersecurity postures, this synergy provides insights into new threats and supports businesses against developing cyber dangers.
Malicious URL Analysis using Machine Learning Algorithm
2024-11-06
573800 byte
Conference paper
Electronic Resource
English
A Method for TLS Malicious Traffic Identification Based on Machine Learning
Trans Tech Publications | 2021
|Detecting Malicious Assembly with Deep Learning
IEEE | 2018
|