Network-based naive Bayes model for social network

Abstract

Naive Bayes (NB) is one of the most popular classification methods.It is particularly useful when the dimension of the predictor is highand data are generated independently.In the meanwhile, social network data arebecoming increasingly accessible, due to the fast development of various social networkservices and websites. By contrast, data generated by a social network are most likelyto be dependent. The dependency is mainly determined by their social networkrelationships. Then, how to extend the classical NB methodto social network data becomes a problem of great interest. To this end, we propose herea network-based naive Bayes (NNB) method, which generalizes the classical NB modelto social network data. The key advantage of the NNB method is that it takes the networkrelationships into consideration. The computational efficiency makes the NNB method even feasible in large scale social networks. The statisticalproperties of the NNB model are theoretically investigated.Simulation studies have been conducted to demonstrate its finite sample performance.A real data example is also analyzed for illustration purpose.