Intelligent Penetration Technology of Power Web System Vulnerability Based on Deep Learning

Abstract

At present, there are some security risks existing in power web system, such as large number of systems, low efficiency of vulnerability identification, poor intelligence level of vulnerability penetration and so on. To solve the problems, this paper studies parallel crawler and multithreaded crawler scanning technology to effectively improve the code and data crawling speed of power web system, so as to improve the efficiency of vulnerability scanning identification. Furthermore, the paper studies the decision-making selection technology of vulnerability feature pattern recognition and vulnerability intelligent detection model of power web system based on LSTM, breaks through the key technologies of semi-automatic outlier sample detection and intelligent vulnerability location and identification, effectively improves the efficiency of vulnerability location and identification, and reduces the labor cost in the process of data processing. Then, based on neural network algorithm, combined with expert experience and exploitation characteristics, the combination rules of parallel and chain are trained. Finally, the deep neural network algorithm is used to judge the feasibility of vulnerability exploitation path, eliminate those paths that cannot be successfully attacked, and improve the success of vulnerability exploitation, so as to improve the ability of intelligent discovery of vulnerability risks.