Pattern Recognition (7): MATLAB implements Naive Bayes Classifier

This series of articles is edited by cloud Twilight. Please indicate the source for reprinting.

Http://blog.csdn.net/lyunduanmuxue/article/details/20068781

Thank you for your cooperation!

Today we will introduce a simple and efficient classifier, Naive Bayes classifier ).

I believe that those who have learned probability theory should not be unfamiliar with the name of Bayes, because an important formula in probability theory is named by Bayes. This is the "Bayesian formula ":

Bayesian classifier is developed based on this formula. The naive word is added here because the classifier makes a hypothesis about various types of distribution, that is, data samples of different classes are independent of each other. This assumption is very strong, but does not affect the applicability of Naive Bayes classifier. In 1997, the Domingos and pazzani of Microsoft Research Institute proved that the classifier still showed good performance even if its premise was not true. One explanation of this phenomenon is that the classifier requires fewer training parameters, so it can avoid overfitting ).

Pattern Recognition (7): MATLAB implements Naive Bayes Classifier