Tamper Detection Based on Webpage Change Features

Abstract

With the rapid development of the Network, the webpage tampering has become a problem that can not be ignored. In October 2019, for example, the number of websites tampered with in China reached more than 20,000, with the largest share coming from Beijing, Shandong and Guangdong respectively [1]. This shows that webpage tampering is no longer a mere simple problem, but requires greater attention. In order to find a better way to detect web page tampering, the website tamper-proof technology receives much attention in the area of web security. This paper * proposes a method of webpage tampering detection based on the webpage change features through analyzing the features of webpage changes and the illegal tampering purpose. Webpage changes will be determined before detecting. The detection model is decided by webpage change time, webpage change code amplitude, webpage change frequency, webpage change content location and webpage change content relationship. To be more specific, the detection process includes two training and detection stages, training phase and detection phase. In the training phase, the effective detection evidence only suitable for the webpage can be obtained through using the data set (multiple changes data in the webpage) to train model. In the detection phase, the detection model will detect the particular webpage according to the detection evidence, and then gives the detection result. If the result is misdeclaration, the detection evidence will be retrained. Furthermore, the simulation tampering experiment is designed to verify the feasibility of the detection method. And the optimal number of the experimental webpage changes firstly into the training phase is determined according to the accuracy and recall rate of the test results. After Verifying, the accuracy and recall of the test results were 98.32% and 99.12% respectively. The best number of changes was 55 and the risk value was 1.