A triadic closure and homophily-based recommendation system for online social networks
https://link.springer.com/article/10.1007/s11280-015-0333-5
This paper studies online social networks (OSNs) and the triadic closure of the friendship recommendation system. As all social media networks users know, sites like Twitter and Facebook recommend friends to the user. However, according to this study, recently these recommendations have been made where these users do not know each other from before. Therefore, these sites are, in a way, “forming” friendships. Based on data analysis, it is demonstrated that many of recommendations are based on friends of friends connections and common interests. But in fact there are two types of friendships: homophily and heterophily, where homophily describes friendships with similar interests and heterophily describes friend groups with diverse interests. In my opinion, I believe there is not one type of friendship, as people gravitate towards those who have similarities but also look for those different than themselves to either be introduced to something new.
This resource connects to the topic of triadic closure covered in class. Triadic closure has to do with the likeliness of two people becoming friends if they have a mutual friend. In other words, if A is friends with B and C, B and C are likely to also become friends. This same property is seen in online social networks, as people find new relationships and connections through their friends. These social media sites know that this is the case and therefore help guide along the process through their recommendation systems.