Predicting the scale of information diffusion in social network services

Predicting the scale of information diffusion is an important task for many social network services (SNS) operators and enterprises. In this paper, the authors investigate the effects of two types of indicators, user attributes and social network attributes, and study on the accuracy of predicting the scale of information diffusion, and how to select an appropriate model that can fit real data better and have a higher accuracy. The experimental results show that both user attributes and social network structure attributes have significant effects on the scale of information diffusion. At the same time, three data mining models are constructed in this paper to predict the scale of information diffusion and compare their prediction accuracy. It is found that neural network model performs much better than decision tree and linear regression do.