Open AccessApplication of Convolution Optimization Algorithm Based on Neural Network in Web Attack Test
Author(s) -
Nan Li,
Sheng Qi Guan
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1325/1/012001
Subject(s) - computer science , web service , data mining , algorithm , web server , artificial neural network , network security , process (computing) , adaptability , machine learning , computer network , the internet , world wide web , ecology , biology , operating system
As society enters the process of information development, the demand for information services in various industries is increasing. Various types of software based on Web services have been widely used in various system platforms, and also bring greater security service requirements. All kinds of network hackers will cause serious damage to the attacks on port access information data. Based on the security of Web services, the paper starts with the characteristics of Web attacks, and studies the structure scheme of Web information feature extraction as a vector model, which is brought into the neural network to calculate by convolution optimization algorithm. The feature vector extraction proposed in the paper is a multi-dimensional training method for convolution input layer. It is suitable for information structure changes in different network environments. By setting the nodes of the Web network as elements of neurons, the connection of several neurons for the model matrix is selected to simulate the effect against web attacks. In the test process, the multi-dimensional space of the proposed algorithm is superior to other algorithms, and has the characteristics of adaptability and controllability. The result of the analysis proves the accurate effect of the research content.