Detection of Malicious SQL Injections Using SVM and KNN Algorithms
Sameer Abduljabbar Kadhim Hacham, Osman N. Uçan · 2023
One of the most destructive types of web application attacks is the SQL injection attack, which typically happens when the attacker or attackers alter, remove, read, and copy data from database servers. Confidentiality, integrity, and data availability are just a few of the security factors that can be affected if a SQL injection attack is successful. Structured query language, also known as SQL, is used to represent queries to database management systems. SQL injection attack detection and deterrence is not a new field of study, but it is still important because techniques from other fields can be used to enhance the attack’s detectability. Techniques from machine learning and artificial intelligence have been tried and tested to manage SQL injection attacks, with encouraging outcomes. During the model evaluation process, the K-Nearest Neighbors (KNN) and Support Vector Machine (SVM) models both obtained good accuracy ratings. The SVM model’s accuracy was approximately 99.08%, while the KNN model’s accuracy was roughly 99.01%.