← Back to Publications List

SQL Injection Detection Using a Siamese Deep Learning Model

Students & Supervisors

Student Authors
Abdullah Al Amin
Master of Science in Computer Science, FST
Supervisors
Firoz Ahmed
Professor, Faculty, FST

Abstract

In the era of modern applications, storing the data of web applications has its own security issues. Recently, cyber threats have become increasingly complex, one of which is the SQL injection (SQLi) attack, that have become the most significant threat to web applications. By exploiting vulnerabilities in SQL queries, attackers compromise sensitive data, disrupt services, and manipulate data, which can result in significant financial losses, reputational damage, and legal consequences. Traditional detection methods often fail to adapt to evolving attack strategies. Existing research often prioritizes raw performance metrics, which limits models to specific web applications. The objective of this research is to develop a siamese deep learning model to detect SQL injection effectively that can handle obfuscated SQL injection attacks. A large SQL injection dataset from Kaggle was used to train the Siamese network with a shared Bidirectional LSTM tower with euclidean distance for similarity learning. The siamese deep learning model for SQLi detection uses query similarity instead of keyword matching that achieves high accuracy and robustness. This addresses the existing SQL Injection Security Vulnerabilities and provides a siamese model that removes previous limitations.

Keywords

SQLi SQLi Detection Deep Learning Siamese Model Siamese Networks.

Publication Details

  • DOI: 10.1109/QPAIN69676.2026.11545955
  • Type of Publication:
  • Conference Name: 2026 IEEE 2nd International Conference on Quantum Photonics, Artificial Intelligence & Networking (QPAIN 2026)
  • Date of Conference: 16/04/2026 - 16/04/2026
  • Venue: IT Business Incubator, Chittagong University of Engineering and Technology (CUET), Chattogram, Bangladesh
  • Organizer: IEEE
View on Publisher