Research And Application Of Naive Bayesian Classification Model

Naive Bayes classifier is a simple and effective classification method. Classifying based on Bayes Technology has got more and more attentions in the field of data mining. In order to get rid of the limit of the assumption of independence among attributes of Naive Bayesian Classifier, this thesis makes a study of two Bayesian classifying model, meanwhile, Naive Bayesian Classification Model is applied to help students in selecting specialties direction. The major work of this thesis is described as follow.Based on the evaluation of condition attribute with correlation improves structure of Naive Bayesian Classification Model. On the basis of analyzing the evaluation of condition attribute with correlation and attribute reduction, satisfied attributes reduction set has been given. According to this method, EANBC is proposed. Compared with Naive Bayesian Classification Model, experimental results show EANBC has higher accuracy.Restricted Bayesian Classification Model Based on Strong Attributes extends the structure of Naive Bayesian Classifier. On the basis of analyzing a variant of Bayes theorem and the evaluation of condition attribute with correlation, SANBC is proposed. Compared with Bayesian Classification Model, experimental results show SANBC has higher accuracy.The Bayesian Classification Model is designed to help the students with their selection of appropriate specialties. By constructing the Bayesian Classification Model and using the experience gained by the students in the past in their selection of specialties, students can base their selection of appropriate specialties on their personal knowledge framework, mastery of knowledge in their fields.