Incremental learning method of Bayesian classification combined with feedback information

Owing to insufficiency of the training sets, the performance of the initial classifier is not satisfactory and can not track the users´ needs. To the defect, the paper proposes an incremental learning method of Bayesian Classifier combined with feedback information. To improve representative ability of feedback feature subset, use an improved feature selection method based genetic algorithm to choose the best features from feedback sets to amend classifier. Analyze the performance of the algorithm by experiments. Experimental results show the algorithm optimizes the classification effect significantly and show better overall stability.