Leveraging Big Data Analytics with Machine Learning to Address Cybersecurity Challenges in ERP Frameworks

Abstract

As digital business systems evolve, there is a rapid surge in infrastructure and data utilization, developing the concept of big data. This big data infrastructure in the enterprise's system premises has led to a tremendous surge in the unabated movement of data across the enterprise for distinctive functionalities. One such impactful business system within the enterprise framework is Enterprise Resource Planning applications, which deal with the creation and maintenance of data. Therefore, this increasing movement of data gives rise to exploitative cybersecurity threats such as infiltration, ransomware, hacking, and phishing, targeting the data used by ERP applications. It imposes a demand to analyze big data generated from ERP applications to identify whether a huge data footprint has forthcoming threats. One such paradigm in investment is to utilize big data analytics along with machine learning approaches to cater to the needs of enterprise cybersecurity. Hence, the entire proclamation in the proposed manuscript revolves around the above thought and addresses leveraging big data analytics with machine learning to address cybersecurity challenges in ERP frameworks: future vision and research directions through systematic literature reviews.A systematic literature review is conducted, considering literature snapshots from 2010 until 2021, using certain scientific study criteria and methodology. It is identified from the reviews that data analytics, when converged with big datasets, can boost the cybersecurity challenges of detection and protection based on the tri-faceted aspects, namely signature-based, anomaly-based, and behavioral analysis-based. Moreover, the integration of analytics, when supported by machine learning data-driven notions, is fast evolving towards a learning and training-based analytical approach with the enterprise dataset. More data can mean more effective machine learning models for predictors. This scientific research study provides a future vision of utilizing machine learning with big data generated from ERP applications to enhance the cybersecurity surveillance strategy, focusing on the contributors of the software application. The valuable insights available are given to show how the data analytics and machine learning-based approaches of other industrial sectors can act as a promising solution for convergence within the ERP framework. These viewpoints provide a valuable guide for new researchers to think and ignite the spark of carrying out future studies on this prospect.