Differentially Private Recommendation System Based on Community Detection in Social Network Applications

Table 1

Description of symbols.

Symbol	Description
	The modularity, evaluation index of community detection
	, is the sum of the weights of all the links in the network
	The weight between node and node
	The sum of the weights of the edges attached from vertex
	The sum of the weights of the links from to nodes in
	The community.
	An algorithm.
	The result of any query operation
	is a random noise and follows the Laplace distribution of scale parameter
	The sensitivity of
	represents a given practical evaluation function. is data set, represents any legitimate output
	is a social network. V represents the set of users in . . The friendship link denoted by . The set of user attributes is denoted by . .
	The set of attribute categories in ,
	The attribute value, equal to label,
	The similarity of attributes between users
	The accuracy of dependence
	The parameters for evaluating utility, it is close to 1 that means utility better, otherwise, it is close to 0, utility lower