Research On Anonymous Protection Technology Of Track Data Based On Sensitivity

In recent years,relevant departments have analyzed and excavated the trajectory data,and have produced various location-based services that have greatly facilitated people's lives.For example,the function of looking for a nearby friend in the social network such as QQ,WeChat,and the function of the living service of recommending a nearby restaurant on the software such as Eleme,Meituan.In addition,by analyzing the trajectory data of mobile terminal users,the relevant government departments can understand the construction of basic transportation facilities and use this information to update and optimize the transportation facilities.However,the trajectory data is closely related to user's daily activities.While enjoying these conveniences,we will also revealing our location information,including sensitive locations information that we do not want to be known.Especially when attackers master these sensitive information and use it.It will lead to serious consequences and even threaten the personal safety of user.In view of the above,based on the in-depth study of cluster-based trajectory privacy protection technology,this paper proposes a(k,?,a_i)-anonymous model based on sensitivity classification.The model can improve the user's privacy while ensuring the availability of the anonymized trajectory data.The main contents of the paper are as follows:First of all,the paper briefly describes the components and system structure of Location Based Service.Then this paper mainly introduces five kinds of privacy protection technologies(generalization,suppression,pseudo trajectories,hybrid region,and clustering),on-line and off-line trajectory privacy protection system structures.Otherwise we analyze their own characteristics,advantages and disadvantages.Besides,the paper deeply studies the(k,?)-anonymous model.And the paper analyzes its advantages and disadvantages.For the(k,?)-anonymous model,which does not consider the privacy disclosure caused by the sensitive differences between different locations,this paper proposes the concept of sensitivity classification,an(k,?,a_i)-anonymous model based on sensitivity classification and SL-NWA algorithm.In addition,the paper introduces two methods of privacy assessment,security and availability,which are often used in the field of trajectory privacy protection.Among them,availability includes information distortion and query distortion rate.Furthermore,the paper proposes a method of security evaluation method that can distinguish between different sensitive locations.Finally,experimental analysis proves that the proposed method can improve the data security while ensuring the data availability.As a result,it improves the quality of the anonymous trajectory data set.