Classification

Supervised and semi-supervised learning algorithms for binary and multiclass problems

Classification is a type of supervised machine learning in which an algorithm “learns” to classify new observations from examples of labeled data. To explore classification models interactively, use the Classification Learner app. For greater flexibility, you can pass predictor or feature data with corresponding responses or labels to an algorithm-fitting function in the command-line interface.

To train regression models, such as logistic regression, regression trees, Gaussian process regression, and support vector regression, see Regression.

Classification Basics

Featured Examples

Identify Noisy Labels Using Confident Learning

Remove mislabeled observations from the training data using confident learning to improve the performance of a classification model.

Open Live Script

Automated Classifier Selection with Bayesian and ASHA Optimization

Use fitcauto to automatically try a selection of classification model types with different hyperparameter values, given training predictor and response data.

Open Live Script

Moving Towards Automating Model Selection Using Bayesian Optimization

Build multiple classification models, optimize their hyperparameters, and select the model that performs the best on a test data set.

Open Live Script

Combine Heterogeneous Models into Stacked Ensemble

Build multiple machine learning models for a given training data set, and then combine the models using a technique called stacking to improve the accuracy on a test data set compared to the accuracy of the individual models.