System and Method for Protecting User Privacy Using Social Inference Protection Techniques

Abstract

A system and method for protecting user privacy using social inference protection techniques is provided. The system executes a plurality of software modules which model of background knowledge associated with one or more users of the mobile computing devices; estimate information entropy of a user attribute which could include identity, location, profile information, etc.; utilize the information entropy models to predict the social inference risk; and minimize privacy risks by taking a protective action after detecting a high risk.

What is claimed is: 
     
         1 . A system for protecting individual privacy in a computer network, comprising:
 first means for modeling a context associated with an individual user and storing the modeled context in a data store;   second means for calculating an information entropy level associated with a user and storing the calculated information entropy level in the data store;   third means for calculating a privacy threshold associated with a user and storing the calculated privacy threshold in the data store; and   fourth means for executing at least one privacy protection action based upon the modeled context, the calculated information entropy level, and the calculated privacy threshold.   
     
     
         2 . The system of  claim 1 , wherein the first means implements a deterministic model of background information associated with a user. 
     
     
         3 . The system of  claim 1 , wherein the first means implements a probabilistic model of background information associated with a user. 
     
     
         4 . The system of  claim 1 , wherein the first means models vicinity information about a user's vicinity. 
     
     
         5 . The system of  claim 4 , wherein the modeled vicinity information includes at least one of names of nearby persons, profiles of nearby persons, and information about nearby locations. 
     
     
         6 . The system of  claim 1 , wherein the first means models personal information about people nearby a user. 
     
     
         7 . The system of  claim 6 , wherein the personal information includes at least one of a user's demographic information, publicly-available information about people, gender information, ethnicity information, geotemporal routines, and individual interests/attributes. 
     
     
         8 . The system of  claim 1 , wherein the second means implements an instantaneous entropy model. 
     
     
         9 . The system of  claim 8 , wherein the instantaneous entropy model models instantaneous information entropy. 
     
     
         10 . The system of  claim 8 , wherein the instantaneous entropy model models instantaneous identity entropy. 
     
     
         11 . The system of  claim 1 , wherein the second means implements a historical entropy model. 
     
     
         12 . The system of  claim 11 , wherein the historical entropy model models historical information entropy. 
     
     
         13 . The system of  claim 11 , wherein the historical entropy model models historical identity entropy. 
     
     
         14 . The system of  claim 1 , wherein the third means determines and stores information about at least one of privacy preferences, anonymity preferences, group privacy preferences, system administrator settings, legal requirements, or social customs. 
     
     
         15 . The system of  claim 1 , wherein the privacy protection action implemented by the fourth means includes at least one of blurring an answer to a user query for information, rejecting an answer to a user query for information, alerting a user as to a privacy risk, informing the user about a current entropy level, informing the user about a history of revealed information, reminding the user about current privacy settings, adjusting the user's privacy settings, and adjusting system administration policy settings. 
     
     
         16 . The system of  claim 1 , wherein the second means implements an inference function of 
       
         
           
             
               
                 
                   INF 
                    
                   
                       
                   
                    
                   1 
                    
                   
                     ( 
                     
                       Q 
                       → 
                       Φ 
                     
                     ) 
                   
                 
                 = 
                 
                   
                     
                       H 
                       max 
                     
                     - 
                     
                       H 
                       c 
                     
                   
                   
                     H 
                     max 
                   
                 
               
               , 
             
           
         
       
       where H max  represents a maximum entropy value and H c  represents a current entropy value. 
     
     
         17 . The system of  claim 16 , wherein the second means implements an inference function of 
       
         
           
             
               
                 H 
                 c 
               
               = 
               
                 - 
                 
                   
                     ∑ 
                     1 
                     V 
                   
                    
                   
                       
                   
                    
                   
                     P 
                      
                     
                         
                     
                      
                     
                       1 
                       · 
                       
                         log 
                         2 
                       
                     
                      
                     P 
                      
                     
                         
                     
                      
                     1 
                   
                 
               
             
           
         
       
       where V is a number of entities having an attribute falling within a pre-defined sphere of influence, and P1 is a probability of a correct inference.