Hi there ...
now a most of researchers are choosing Nave-Bayes Method As Base Method for Handling the uncertain reasoning in some of expert systems while this method is a weak learner because of the additional assumptions that added into the Bayes method .
So why researchers choosing this method ? or why they do to this method to increase the efficiency ?
thank's for your help .
Naiev Bayes method is based on probability therefore it is easy to implement and with little time and space complexity.
Naive Bayes classifiers are highly scalable, requiring a number of parameters linear in the number of variables (features/predictors) in a learning problem.
Naiev Bayes method is based on probability therefore it is easy to implement and with little time and space complexity.
Naive Bayes classifiers are highly scalable, requiring a number of parameters linear in the number of variables (features/predictors) in a learning problem.
Thank you Hussam Ali .
The properties of Naive Bayes that you said is came from some assumptions that added to classic Bayes classifier that make Naive Bayes far from the real word problems and that is why several criticized was against of this method .
But after all of this ,some researchers are choose this method to handle uncertainty.
The question is how they make Naive Bayes to work good in real word problem ?
I mean how they could increase performance of this method?
For Text data check this paper
http://people.csail.mit.edu/jrennie/papers/icml03-nb.pdf
For what kind of data you want to use Naive Bayes?
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Artificial Intelligence