| Title | Efficient Detection Of SQL Injection Attack(SQLIA) Using Pattern-based Neural Network Model |
| Publication Type | Conference Paper |
| Year of Publication | 2021 |
| Authors | A, Meharaj Begum, Arock, Michael |
| Conference Name | 2021 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS) |
| Date Published | feb |
| Keywords | composability, Computational modeling, Cross Layer Security, cyber attack, MLP, parsing, Payloads, phishing, pubcrawl, Resiliency, security, SQL Injection, SQLIA, Structured Query Language, Tagged Patterns, tagging |
| Abstract | Web application vulnerability is one of the major causes of cyber attacks. Cyber criminals exploit these vulnerabilities to inject malicious commands to the unsanitized user input in order to bypass authentication of the database through some cyber-attack techniques like cross site scripting (XSS), phishing, Structured Query Language Injection Attack (SQLIA), malware etc., Although many research works have been conducted to resolve the above mentioned attacks, only few challenges with respect to SQLIA could be resolved. Ensuring security against complete set of malicious payloads are extremely complicated and demanding. It requires appropriate classification of legitimate and injected SQL commands. The existing approaches dealt with limited set of signatures, keywords and symbols of SQL queries to identify the injected queries. This work focuses on extracting SQL injection patterns with the help of existing parsing and tagging techniques. Pattern-based tags are trained and modeled using Multi-layer Perceptron which significantly performs well in classification of queries with accuracy of 94.4% which is better than the existing approaches. |
| DOI | 10.1109/ICCCIS51004.2021.9397066 |
| Citation Key | a_efficient_2021 |
Efficient Detection Of SQL Injection Attack(SQLIA) Using Pattern-based Neural Network Model