Scholarlink

The web is a critical application in a variety of online services, including e-banking and e-commerce. However, the potential threat of SQL Injection (SQLI) attacks to sensitive data that is processed online is on the rise. SQLI can be employed to compromise confidential information from the targeted database and obtain unauthorized access by exploiting vulnerabilities in online applications. This paper introduces a detection system that is powered by machine learning and is designed to mitigate such attacks. A hybrid system that integrates Slime Mold Algorithms (SMA) and Support Vector Machine (SVM) was proposed as a novel method for safeguarding web applications from SQLI attacks. The SMA algorithm is employed in this paper to determine the most effective values for SVM parameters to enhance performance. To evaluate the proposed system performance, we compiled a dataset including normal and attack SQL queries In addition to that Extracted a set of statistical features that perfectly describe each query. The experimental results show that the proposed method SMA-SVM obtained 0.9931 concluding that this method effectively detects SQL injection attacks rather than compared to single-SVM.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Support Vector Machines and Slime Mold Optimization Algorithms for SQL Injection Detection

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Zainab H. Al-Araji,
  • Hasanen Alyasiri

摘要

The web is a critical application in a variety of online services, including e-banking and e-commerce. However, the potential threat of SQL Injection (SQLI) attacks to sensitive data that is processed online is on the rise. SQLI can be employed to compromise confidential information from the targeted database and obtain unauthorized access by exploiting vulnerabilities in online applications. This paper introduces a detection system that is powered by machine learning and is designed to mitigate such attacks. A hybrid system that integrates Slime Mold Algorithms (SMA) and Support Vector Machine (SVM) was proposed as a novel method for safeguarding web applications from SQLI attacks. The SMA algorithm is employed in this paper to determine the most effective values for SVM parameters to enhance performance. To evaluate the proposed system performance, we compiled a dataset including normal and attack SQL queries In addition to that Extracted a set of statistical features that perfectly describe each query. The experimental results show that the proposed method SMA-SVM obtained 0.9931 concluding that this method effectively detects SQL injection attacks rather than compared to single-SVM.