Beyond friendship: the art, science and applications of recommending people to people in social networks

摘要: While Recommender Systems are powerful drivers of engagement and transactional utility in social networks, People recommenders a fairly involved diverse subdomain. Consider that movies recommended to be watched, news is read, people however, for plethora reasons -- such as recommendation befriend, follow, partner, targets an advertisement or service, recruiting, partnering romantically join thematic interest groups.This tutorial aims first describe the problem domain, touch upon classical approaches like link analysis collaborative filtering then take rapid deep dive into unique aspects this space reciprocity, intent understanding recommender recomendee, contextual recommendations communication flows referrals paradigm delivery using graph. These will discussed context published original work developed by authors their collaborators many cases deployed massive-scale real world applications on professional networks LinkedIn.