People tend to build and maintain their friendship relying on SNS nowadays. Thus, the problem of how to organize the social network accurately and automatically. In this paper, a hybrid neuro-fuzzy approach is used.
. Many aspect impact the error values like input/output, membership functions, the training data arrays, and the number of epochs needed to train the model. This paper is based on hybrid Neuro-Fuzzy concept for testing the link prediction for facebook data. We use Matlab to calculate average testing Error, View Generation Rule, OutputSurface.