A Progressive Machine Learning Algorithm for Detecting Intrusions in Corporate Cloud Applications

An Intrusion Detection System (IDS) is a sophisticated software application designed to monitor network traffic for signs of fraud or violations of web application policies. It usually includes a security data management system that alerts users to potential threats or policy breaches. This study focuses on creating an innovative approach to intrusion detection in financial cloud applications, leveraging machine learning (ML) techniques and utilizing log data. We introduce a unique method that integrates CatBoost with the Chameleon Optimization Algorithm (CC-COA) to detect attacks on web and cloud-based applications within the financial sector. CatBoost was chosen for its effectiveness in managing categorical data, while COA was employed for fine-tuning hyper parameters. The model’s performance is assessed using various metrics, including accuracy, precision, F1 Score, and recall. The results of our experiments indicate that the proposed model significantly surpasses traditional methods in detecting different types of attacks in cloud-based financial environments. Our model was built using log files gathered from the websites of several leading local banks. Keywords: Log-Based Intrusion Detection, Financial Cloud, Machine Learning (ML), Integration of CatBoost with Chameleon Optimization Algorithm (CC-COA), Cloud Data Security.