Detecting the Abnormal SQL Query using Hybrid SVM Classification Technique in Web Application

Authors

  • R. Shobana
  • M. Suriakala

Abstract

SQL Injection Attacks (SQLIAs) are playing a significant role in database driven sites due to its automatic nature. Previously, many works had been carried out to reduce this SQLIAs at the application side but, they result in failure in many ways. Many techniques were tended to minimize the usage of less number of support vectors. In this paper, the proposed methodology will be fully concentrate on minimizing the dataset points and that leads to improvement of SVM classification. The main idea is to calculate the approximate rate of decision boundary rate of decision boundary of SVM by the assistance of binary trees. The finally obtained tree is considered as the hybrid tree that will helps to sense both of theuni-variant and multi-variant nodes. The hybrid tree takes SVM's assistance just in ordering significant information focuses lying close choice limit, staying less urgent datapoints are grouped by quick uni-variant nodes.

Downloads

Published

2020-04-09

Issue

Section

Articles