Intrusion Detection System for Cloud ERP’s using ML Techniques

Kalidindi, Vinay (2024) Intrusion Detection System for Cloud ERP’s using ML Techniques. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (762kB) | Preview
Preview	PDF (Configuration Manual) Download (535kB) | Preview

Abstract

This paper argues that the evolution of ERP systems in cloud environments has increased the flexibility and availability of business processes. But it causes new risks for security problems: unauthorized access, data leaks, and cyber threats are favored by the nature of the cloud environment being open and shared. These make the integration of Intrusion Detection Systems (IDS) with cloud-based ERPs a crucial area of research owing to the arises of challenges. IDS are important for creating filters for monitoring the network traffic and timely detection presence of intruders to offer a protective layer for the network against intrusion. This project uses Azure Machine Learning Notebook for constituting the different ML models, such as Random Forest, Logistic Regression, and Support vector machine for building, training, and testing. The Random Forest classifier achieved an accuracy of 99.94%, precision of 99.97%, and recall of 99.88%, making it highly effective in distinguishing between normal and malicious network traffic. By integrating Microsoft Forms, Power Automate, and Azure Machine Learning, the system ensures real-time detection and automation for ERP-like scenarios. This study offers practical insights into improving ERP system security through scalable and adaptive ML solutions. The models were built based on the pre-processed KDD Cup 1999 data set which is considered to enclose all kinds of network activities and anomalous events. An application user interface was created using Stream lit which allowed the model to predict and monitor intrusion attempts in real-time. To mimic the functionality of an ERP system, several forms were created while Microsoft Power Automate was used to automate the selected workflows, which in their turn would generate emails upon intrusions. By combining these technologies, the IDS continuously tracks the ERP like activity, analyses inherent vulnerabilities, and informs the administrator regarding threats, making the IDS active in providing security. This project shows that it is possible to integrate machine learning and automation with cloud tools to improve ERP system security and create novel solutions based on them for enterprise environments. The recommendations drawn from this research are useful in enhancing basic protection and strengthening of ERP systems based on cloud.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Jaswal, Shivani UNSPECIFIED
Uncontrolled Keywords:	Intrusion Detection System (IDS); Machine Learning (ML); Cloud-based ERP; Random Forest; Anomaly Detection
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Cloud computing Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Cloud Computing
Depositing User:	Ciara O'Brien
Date Deposited:	15 Jul 2025 13:12
Last Modified:	15 Jul 2025 13:12
URI:	https://norma.ncirl.ie/id/eprint/8111

Actions (login required)

View Item