Investigating Supervised Machine Learning Methodologies for Preventing Phishing Attacks on SCADA Server

A high number of industries utilize supervisory control and data acquisition system for controlling and monitoring day-to-day operations related to the organization’s domain with digital methods and accuracy to boost the overall infrastructure. These organizations are extensively involved in sending and receiving crucial data from multiple nodes. These systems can become the target of Phishing Attacks that is used to steal the private data of the user. Attackers steal the user's important credentials like user login and password involved in these organizations. The SCADA has seen an increase in Phishing activities over the past ten years as well. To overcome these phishing attacks machine learning techniques can be trained in a prior manner for detecting these attacks in the organization. Machine Learning algorithms can examine the details of each packet by analysing traffic patterns. In this research paper, the proposed model works on preventing SCADA server attacks using artificial intelligence-enabled machines at an advanced stage. The more related three machine learning algorithms are explored for the classification of data packet and these are decision tree, k-nearest neighbors and linear regression for detecting Phishing Attacks on the SCADA system. The dataset included in the experimental setup is the interpretation of data packet traffic features. This paper demonstrates the promise of these algorithms for more research in this field and offers interesting details on the application of machine learning in spotting dishonest conduct inside SCADA phishing attempts.