Research And Implementation Of Security Analysis For Mobile Social Users

With the rapid development of mobile Internet,mobile social networking has become an important way for people to communicate with each other.The domestic mobile social network represented by mobile QQ and WeChat has quickly become the most popular social mode nowadays due to its characteristics,such as mobility,convenience and sociality.However,with the continuous expansion of mobile social users and the relative concealment and diffusion effect of mobile social networks,mobile social networks are increasingly being targeted by malicious users and have become hotbeds of various types of unsafe incidents.Therefore,it is of great practical significance to supervise and manage the mobile social networks and create a healthy and safe mobile social networking environment for the users.In social network supervision and management,social user security analysis technology has become a research hotspot in recent years.At present,the researchers on security analysis of social users at home and abroad focus on two aspects,one is based on user's social content,including malicious information content and related statistical characteristics,on the other hand,based on social user behavior characteristics and social user relationship.Relevant experts and scholars usually start from one or two aspects mentioned above to analyze the social users on the online social platform and obtain great research results.However,in mobile social networks,because the user's social account and social content usually contain more privacy information,so the user privacy protection strength is higher,which results in the inability to utilize the existing web crawler technology and the open API of the social platform to collect mobile social user data,such as social content,behavioral characteristics and user relationships,thus can not analyze the security of mobile social users directly.On the other hand,the current supervision and management of mobile social users mainly relies on user reporting,which is usually handled after malicious behavior occurs,or even after a certain scale is formed.Such post-tracing methods are difficult and time-consuming to obtain evidence.Therefore,it is of great research value to solve the problem of data collection for mobile users,to analyze the security of mobile social users before the malicious behavior occurs or before it is formed,and to identify the malicious users.In this paper,we do a research on mobile social network,about its basic concepts,the characteristics of information dissemination as well as social models.Then we propose a security analysis method of mobile social user based on the traditional social user security analysis combining with the characteristics of Android application interface structure and Android automated testing technology.The method firstly studies the data collection method of mobile social user,design and implements an Android-based automatic data acquisition tool which can collect mobile social user data,such as social group data and user data.The use of this data acquisition tool can overcome the acquisition problems of mobile social user data.Then,through systematic research on social groups,we propose to combine the conversation flow of social groups with the group description information,and use natural language processing techniques to mine the topics of these two parts so as to find out the groups whose conversation flows and group description are mismatched.After finding abnormal social groups,we further combines the content characteristics with behavioral characteristics of group members to identify malicious users.Finally,we do experiments to prove that the mobile social user data acquisition tool designed in this paper can effectively collect user data,and verify that the proposed method for mobile social user security analysis has good accuracy and feasibility.