Machine Learning: 2 Books in 1: Machine Learning for Beginners, Machine Learning Mathematics. An Introduction Guide to Understand Data Science Through the Business Application

Types of "Naïve Bayes classifier" 
“Multinomial Naïve Bayes” - This is widely used to classify
documents, for example, which category does a document
belong: 
beauty, 
technology, 
politics, 
and 
so 
on.
The frequency of the phrases in the document is considered as the
features or predictors of the classifier.

---

“Bernoulli Naive Bayes” - This is nearly identical to the
“Multinomial Naïve Bayes”, however, the predictors used here
are the "boolean variables". For example, depending on whether
a select phrase occurs in the text or not, the parameters used to
predict the class variable can either be a yes or no value.
“Gaussian Naive Bayes” - When the predictors are not distinct
and have very similar or continuous values, it can
be assumed that these values are obtained from a Gaussian
distribution.
Applications of “Naïve Bayes”
“Real-Time Prediction”: Naive Bayes is extremely quick in
learning from the input data and can be seamlessly used to
generate predictions in real-time.
“Multi-class Prediction”: This algorithm is also widely used to
generate predictions for multiple classes at the same time. It
allows the prediction of the probability of various classes of the
target variable.
“Text classification / Spam Filtering / Sentiment Analysis”:
The “Naive Bayes classifiers” is heavily used in text
classification models owing to its ability to address problems
with multiple classes of the target variable and the rule of
autonomy. This algorithm has reported higher success rates than
any other algorithm. As a consequence, it is commonly used in
for identification of spam emails and sentiment analysis

---

by identifying favorable and negative consumer feelings on
the social media platforms.