Network Embedding For Link Prediction in Bipartite Networks

Many social networks have a bipartite nature. Link prediction in has been the focus of interest for many researchers recently. Network embedding, which maps each node network to low-dimensional feature vector is used solve problems. The aim this study investigate how embedding enhance link performance networks. A and supervised learning based model presented input learned vectors pairs obtained from method. target binary label indicating existence or absence between these pairs. Ensemble algorithms applied prediction. experiments performed on two built public datasets led promising results with 0.939 0.974 AUC values. Random Forest models trained BiNE method achieved highest performances.