Abstract
One of the most destructive types of web application attacks is the SQL injection attack, which typically happens when attackers modify, add, delete, retrieve, and try to copy sensitive data from a database. Confidentiality, integrity, and data availability are just a few of the security factors that are impacted by a successful SQL injection attack. Queries are used in database management systems which are represented using SQL (structured query language). Using inputs that can confuse the SQL compiler is a way that attackers perform this attack some variations of attacks are tautologies, illegal/logically incorrect queries, Error-based SQL injection, Union queries, piggy bank queries, and Blind injection.
There are multiple ways to combat SQL injection, such as analyzing database inputs, restriction of the database code, checking on the data before it is sent to the user, restricting database access to low-level users, maintaining databases, and Monitoring applications these techniques can minimize the SQL injection but cannot wholly prevent it. To add more security to the websites adding Deep learning models that can detect SQL injection would be an important asset.
The primary objective of this project is to establish an online platform designed to accommodate various deep learning and machine learning models that can detect SQL injection. This platform facilitates application developers in obtaining Machine Learning and Deep Learning models, allowing them to enhance the security of their applications by utilizing these models to forecast whether SQL injection is occurring. The developers can use this forecast to prevent SQL injection.