NETWORK'S ACCESS LOG CLASSIFICATION FOR DETECTING SQL INJECTION ATTACKS WITH THE LSTM ALGORITHM

SQL Injection attacks are one of the popular web attacks. This attack is a network security problem focused on the application layer which is one of the causes of a large number of user data leaks. Currently available SQL detection techniques mostly rely on manually created features. Generally, the detection results of SQL Injection attacks depend on the accuracy of feature extraction, so they cannot overcome increasingly complex SQL Injection attacks on various systems. Responding to these problems, this research proposes a SQL Injection attack detection method using the long short term memory (LSTM) algorithm. The LSTM algorithm can learn data characteristics effectively and has strong advantages in sorting data so that it can handle massive, high-dimensional data. The research results show that the accuracy of the model approach created is able to recognize objects with a high accuracy value of 98% in identifying SQL Injection attacks.