AI-Enabled Approach for Preventing DNS Attacks on Banking Institutions

In today’s era of digitalization financial institutions like banks, insurance companies are growing and migrating for complete automation process. Various financial activities like money transfer, money deposit, share market, bill payment and many other tasks related to transactions can be performed using web based or mobile based applications. Due to the heavy use of online portals, security of banking servers is one of the major concerns to maintain privacy and keep the data secure from unauthorized access. Domain Name System attacks are widely executed on banking sector by hackers for stealing the private information of users or bank officials. Machine Learning training for introducing security and trust in the system can lead to resolve these issues. In this paper, the model is proposed to protect the banking institutions from domain name system attacks using Machine Learning trained machine, where this machine can decide whether network packet should enter the banking network or not. Machine Learning algorithms are used to find the attacks in the real-time environment and the algorithms are Random Forest classifier, Logistic Regression, Support Vector Machine, and Gaussian Naive Bayes. After applying this algorithm Support Vector Machine outperforms in comparison with other techniques for detecting domain name server attacks. The consequences are linked based on the metric accuracy, precision, F1 score and recall for the early detection of cyberattacks on DNS bank server.