An approach of anchor link prediction using graph attention mechanism

Nowadays social networks such as Twitter, LinkedIn, and Facebook are a popular necessary platform. It is considered miniature of an actual network because its advantages in connecting sharing information between users. The analysis data on online has become field that attracted lot attention from the research community anchor link prediction one main directions this field. Depending demand, user can simultaneously participate many different networks, kind task determines identity networks. In article, we proposed algorithm missing/future links users two Our utilizes graph technique to represent source target into low-dimension embedding spaces, then apply canonical correlation recline their embeddings same latent spaces for final prediction.