Ranger Random Forest-Based Efficient Ensemble Learning Approach for Detecting Malicious URLs

Abstract

The massive quantity of data generated from a variety of resources is hard to manage the process and analyze. Big data enables machine learning algorithms to find meaningful patterns and make accurate predictions. Big data analytics can be used in organizations to process and analyze the vast amount of data to find the insights of data, such as risks, threats, and incidents. These incidents create more security issues. A wide range of non-constructional activities happening in WWW requires the detection of malicious URLs for internet security. Malicious URLs or Websites are hosting spontaneous content and involve many users to become victims of scams. This paper presents ensemble learning-based, faster, and memory-efficient random forest algorithm for detecting malicious URLs. The proposed method proves scale best with the number of instances and variables.