Semantic based trust recommendation system for social networks using virtual groups

Billions of people create trillions of relations through social media each day, but only few of us consider how each click and key press frames relationships that, in aggregate, form a vast social network. Existing social networks recommend friends to users based on their social graphs like mutual friends, which may not be the most appropriate to reflect a user's taste on friend selection in real life. In order to improve the existing methods, this paper propose a semantic based trust recommendation system which recommend trust companion having high similarities in message sharing. Positive and negative approaches towards the shared posts or links can be determined by likes and dislikes and also consider the opinions towards comments. As a result, the semantic based recommendation system maintains virtual group of trust people having similar interest. On line virtual groups are becoming progressively prominent due to the growth of community and social networking sites. The proposed framework is helpful for handling trust peoples in social networks, which is based on a prominence mechanism that captures the connections between the network members(trust user, distrust user), analyzes the semantics of these connections, and provides personalized user recommendations.