What is the difference between dynamic Naive Bayes Classifier and Naive Bayes Classifier

Question

I current use HMM to predict the solar radiation.

The traditional HMM model can only take one observation sequence data into consideration.

However, when I use Gibbs sampling to estimate the hidden states reduces 4 and error is high.

Now, I am thinking how to input more than one piece of data into the model like use temperature and consumption data both as observations.

So I search for the Dynamic Naive Bayes Classifier. However, I do not know what is the Dynamic Naive Bayes Classifier and how to implement it.

Could someone give me some answers or provide me some tutorials.

Thanks in advance :)

Answer 1

The naive Bayesian classifier is a supervised machine learning model used to perform the classification task for the given set of training and testing data with an assumption that all features are independent for the assigned class labels. On the contrary, Dynamic Naive Bayesian classifiers is a generalized version of the HMM model that can model the multivariate observation sequences. For more details, please see the following paper, you only need to change some part of the HMM mode to get your own DNBC classifier.

1. http://www.scielo.org.mx/scielo.php?script=sci_arttext&pid=S1665-64232011000100007

Thanks