Research On Malicious URL Detection For Mobile Internet

With the rapid development of mobile Internet and the rapid spread of mobile terminals,the problem of network security has become increasingly serious.At present,users are facing more and more threats such as malicious URL attacks and fraud.The effective method of detecting malicious URLs is very important to reduce mobile Internet crimes,reduce the economic losses of netizens,and create a good Internet environment.At the same time,it is also a very worthwhile research hotspot in the field of network security.The focus of this study is how to apply the Internet's detection method of malicious URLs to mobile networks.At the same time,the use of the characteristics of the mobile Internet itself to further improve the detection of malicious URLs.Based on this background,this paper first studies the detection methods and limitations of the existing malicious URLs on the Internet,and the detection of pseudo-base stations with mobile network-specific scenarios that spread malicious URLs.In addition,it describes in detail the related technologies of offensive and defensive URLs in mobile networks,and details Mainstream machine learning classification algorithms for the detection of malicious URL static text features are described.Through the comparison of the analysis of the research status and the introduction of various machine learning algorithms,this paper firstly uses the convolutional neural network with automatic feature extraction function,which is a deep learning algorithm for malicious URL detection of static features.This algorithm can get rid of the Shallow machine learning takes time and effort to detect and process malicious signatures.In order to allow the algorithm to have more detailed feature extraction,this paper uses not the word-level embedded vectorization but the character-level embedded vector.For semantic recognition.At the same time,an algorithm for detecting malicious URLs by detecting pseudo-base station behavior characteristics is proposed.By analyzing the behavior of pseudo-base stations in detail,a malicious URL model based on pseudobase station behavior detection is established to optimize the model from the perspective of parameter selection algorithm.Finally,using the real mobile network data as experimental samples,it is applied to the two malicious URL models designed in this paper.The experimental results are compared and analyzed to verify the validity of the above model.