Machine Learning Blunts the Needle of Advanced SQL Injections

Volkova, Marina; Chmelar, Petr; Sobotka, Lukas

Machine Learning Blunts the Needle of Advanced SQL Injections

Files

2019-1-023.pdf(1007.55 KB)

Date

2019-06-24

Authors

Volkova, Marina

Chmelar, Petr

Sobotka, Lukas

Publisher

Institute of Automation and Computer Science, Brno University of Technology

Altmetrics

Abstract

SQL injection is one of the most popular and serious information security threats. By exploiting database vulnerabilities, attackers may get access to sensitive data or enable compromised computers to conduct further network attacks. Our research is focused on applying machine learning approaches for identication of injection characteristics in the HTTP query string. We compare results from Rule-based Intrusion Detection System, Support Vector Machines, Multilayer Perceptron, Neural Network with Dropout layers, and Deep Sequential Models (Long Short-Term Memory, and Gated Recurrent Units) using multiple string analysis, bag-of-word techniques, and word embedding for query string vectorization. Results proved benets of applying machine learning approach for detection malicious pattern in HTTP query string.

Keywords

SQL injection identication, machine learning, deep learning, recurrent neural networks, text analysis, web application firewall, intrusion detection system

Citation

Mendel. 2018 vol. 25, č. 1, s. 23-30. ISSN 1803-3814
https://mendel-journal.org/index.php/mendel/article/view/74

Document type

Peer-reviewed

Document version

Published version

Language of document

en

Document licence

Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International license
http://creativecommons.org/licenses/by-nc-sa/4.0