Given a directed social graph, have to predict missing links to recommend users (Link Prediction in graph).
Generated training samples of good and bad links from given directed graph and for each link create new features using feature engineering techniques like no of followers, followed back, cosine similarity, Jaccard similarity, page rank, katz score, adar index, some svd fetures of adj matrix, some weight features etc. and trained ML models based on these features to predict link.
https://www.kaggle.com/c/FacebookRecruiting/overview/description