CYBER ATTACKS DETECTION THROUGH MACHINE LEARNING IN BANKING

Abstract

Cyberattacks may cause a wide range of problems, from power outages to broken military equipment to the loss of vital information like patient medical records. Due to the huge monetary worth of the information banks keep, they are a prime target for cybercriminals. The larger the digital footprints of banks, the easier it is for hackers to target them. This study examines the Banking Dataset for indicators of cyber attacks on financial institutions. In this research, CYBER attacks have been predicted using a combination of classification techniques. We have increased the complexity of generic model architecture in an effort to boost their performance. The support vector machine (SVM) was not the only technique we utilized; the k-nearest neighbors (KNN) and random forest (RF) methods were also used. When compared to the KNN and RF, the SVM's detection accuracy of 99.5% was much superior. When compared to KNN, RF, and other established ML/DL techniques, the SVM has been determined to be the most reliable.