A recommender system on social networks according to user interaction using fuzzy clustering and neural network

Social networks are becoming popular recent years. Because the channels that people can communicate quickly via personal electronic devices in any place and at any time with friends and family. after web 2 and spread social network in recent years researchers find another source to provide better quality of recommendation, that was social network data. and planning a lot of researches about this field.For modeling this network s used from graph system that people are nodes and relation between people are edges. In the meantime, there is a large graph; That the clustering is doing using Fuzzy clustering operations. And then using neural network modeling is performed. Clustering in three ways FCM, PCM, FPCM done and the results are compared with each other, the results can offer high accuracy in modeling the relationships within the graph shows. Accurately simulated neural network and fuzzy clustering up to 89 percent improvement, one percent improvement than svm clustering.