Classifying Database Users for Intrusion Prediction and Detection in Data Security

Abstract

The fact that users and applications acquire information using web sites on the internet means that document and information sharing, banking and other operational processes are increasing day by day. Recently however, with the widespread use of the internet, some security problems, such as unauthorized access, data breaches, code infection, malware infections, data leaks and distributed denial of service attacks have emerged. This situation necessitates the protection of the information used in personal and public spaces. In this study, a common model was created to detect user intrusions by taking into account criteria such as the number of transactions performed, their IP addresses, the amount of data they use, the transaction type they perform and the roles they undertake. In this way, the aim was to ensure database security by detecting risky user groups in advance.