Research On Privacy Risk Assessment Of Social Networks

Social networking sites play an important role in modern life,and more and more users are incorporating these social networking sites into their daily activities.The huge increase in the use of social networking sites has led to an explosive growth in online sharing of information,but the correctness and legitimacy of the information can be manipulated.More and more privacy leaks have been reported by the media,and users are paying more and more attention to privacy protection.The behavior of users in social networks is closely related to the privacy of users,which exposes users to major risks such as identity theft,economic loss and even physical harm.Although social networking sites allow users to set their own private information,few users pay attention to this.Therefore,a privacy risk assessment model is needed to assess the risk of user privacy and timely warning to prevent privacy leakage.At present,most of the research on social network privacy focuses on privacy protection.Researchers usually solve users' privacy protection problems from data,social attributes and social relations.However,there are few studies on user privacy discovery and privacy risk assessment in social networks.There is still a lack of uniform standards for which information in social networks can be defined as privacy information.In the research on privacy risk assessment,most researchers only select a small number of angles to evaluate privacy risk,which is difficult to guarantee the comprehensiveness of privacy risk assessment.Therefore,it is an important research direction to evaluate which privacy exists in social networks and evaluate users' privacy risks comprehensively,and to warn users from the source of privacy information.The main contents and innovations of this paper are as follows: 1.The privacy information existing in user data and text content is analyzed,and the privacy system of user data and text content is constructed.2.Establish a model based on AHP and grey theory to evaluate the privacy risk in user data.3.Construct a classifier for privacy text detection,and evaluate the privacy risk in text content based on it.This paper proposes a privacy risk assessment model,which comprehensively evaluates the privacy risk from the user data and text content.At the same time,based on information entropy,a quantification method of text content information privacy is proposed,which can be effectively used for privacy risk assessment.By evaluating the privacy risk of a large number of actual data,the results show that the privacy risk assessment model proposed in this paper can effectively assess the user data privacy risk and text content privacy risk of social network users.This model can be used in the process of evaluating the users' privacy risk in social network,and can warn the users' privacy risk in time,so as to protect the users' privacy security.This study can provide constructive suggestions for the process of privacy discovery and privacy protection.