Special Issue Article Open Access

Inferring Private Information from Social Network Using Collective Classification

Abstract

Online social networks are used by many people. These Social networks allow their users to connect by means of various link types in which the network gives an opportunity for people to list details about themselves that are relevant to the nature of the network. Here there is a chance of inference when user released some personal information in the network. Social network is represented as graph structure in which nodes and edges denotes user’s of network and relationship links with friends. In this paper, the social network data has been classified with the help of collective classification (both node and link classification) method. Using the collective classification method the system could infer more sensitive information from the network with high accuracy. In collective classification method, it involves three components called local classifier, relational classifier and collective inference. From this experiments conducted in this research work, it is observed that the proposed work provide better classification accuracy due to the application of collective classification method in link analysis.

A.Annapoorani, Ms.P.Indira Priya

To read the full article Download Full Article