An Egocentric Algorithms of Extractions of the Interests of the users from the RSN

Abstract

The search for information (RI) personalized tends mainly to model the user according to a profile and then to integrate it into the chain of access to information, in order to better meet his specific needs.

The user profile is a central element in the information adaptation systems. We are interested in the process of enriching the user profile from his social network, which represents a very rich source of information about the user, but the problem that arises is that the user profile may not contain all the interests and information that may be useful for a given mechanism, especially for new users of the system and those who are not very active. In order to solve these problems we will use a mechanism to detect users similar to this user in the system, and analyze their interests, using similarity techniques, and consequently use the CoBSP algorithms by improving its performance in a very remarkable way by extending the user's 1-egocentric network, which allows to add more nodes from different social networks to cover all the interests of the user.

So our proposal is to combine the user's (friend's) relationships on different types of social networks (here Facebook, Twitter and LinkedIn) in a single social egocentric graph before the application of the CoBSP algorithm whose goal is to obtain a more complete User Profile that brings together the majority of the user's interests.