Naive Bayes models assume that observations have some multivariate distribution given class membership, but the predictor or features composing the observation are independent. This framework can accommodate a complete feature set such that an observation is a set of multinomial counts.
To train a naive Bayes model, use fitcnb in the command-line interface. After training, predict labels or estimate posterior probabilities by passing the model and predictor data to predict.
| Classification Learner | Train models to classify data using supervised machine learning |
ClassificationNaiveBayes | Naive Bayes classification for multiclass classification |
CompactClassificationNaiveBayes | Compact naive Bayes classifier for multiclass classification |
ClassificationPartitionedModel | Cross-validated classification model |
Train Naive Bayes Classifiers Using Classification Learner App
Create and compare naive Bayes classifiers, and export trained models to make predictions for new data.
Supervised Learning Workflow and Algorithms
Understand the steps for supervised learning and the characteristics of nonparametric classification and regression functions.
Categorical response data
The naive Bayes classifier is designed for use when predictors are independent of one another within each class, but it appears to work well in practice even when that independence assumption is not valid.
Plot Posterior Classification Probabilities
This example shows how to visualize classification probabilities for the Naive Bayes classification algorithm.
This example shows how to perform classification using discriminant analysis, naive Bayes classifiers, and decision trees.
Visualize Decision Surfaces of Different Classifiers
This example shows how to visualize the decision surface for different classification algorithms.