S. Yuan’s scientific contributions

When real problem domain deals with continuous variables, researchers always apply discretization before modeling. It extends the naive Bayes's conditional independence assumption to handle continuous variables and then introduces a new learning algorithm-general naive Bayes (GNB), which can not only avoid the negative effect of discretization on prediction accuracy, but also apply incremental learning to make full use of the information from data. Experimental results on a variety of UCI data sets suggest great improvement from the viewpoint of prediction accuracy and independence assumption, respectively.