Abstract:

Nowadays Online Social Networks have gained the most popularity among different communication platforms, and an extensive applications variety is measured for them. Since with the increase in the popularity of Online Social Network, friendship is one of the essential connections amongst members of a social network. Matching of friends is a practical technique to recommend people to each other. In this paper centrality measures are used in Friends of Friends network and similarity measures in group and events network to recommend friend. The Facebook 2010 dataset was used which include the data of friends, groups and events of different users. Friends are recommended based on two different techniques. The first technique is based on friend of friends where friends are recommended based on mutual friends and centrality measure. The second technique used similarity measures to recommend the friend. Lastly, the comparative analysis of both the techniques was done by evaluating accuracies of each technique. The accuracy was calculating by comparing 2010 dataset with the 2018 dataset.