Naive Bayes

../../_images/naive-bayes.png

Naive Bayesian Learner

Signals

Inputs:

  • Data

    A data set

  • Preprocessor

    Preprocessed data

Outputs:

  • Learner

    A Naive Bayesian learning algorithm with settings as specified in the dialog. It can be fed into widgets for testing learners.

  • Naive Bayesian Classifier

    A trained classifier (a subtype of Classifier). The Naive Bayesian Classifier signal sends data only if the learning data (signal Data) is present.

Description

../../_images/NaiveBayes.png

This widget has two options: the name under which it will appear in other widgets and producing a report. The default name is Naive Bayes. When you change it, you need to press Apply.

Examples

Here, we present two uses of this widget. First, we compare the results of the Naive Bayesian learner with another learner, the Random Forest.

../../_images/NaiveBayes-Predictions.png

The second schema shows the quality of predictions made with Naive Bayes. We feed the Test&Score widget a Naive Bayes learner and then send the data to the Confusion Matrix. In this widget, we select the misclassified instances and show them in Scatterplot. The bold dots in the scatterplot are the misclassified instances from Naive Bayes.

../../_images/NaiveBayes-Misclassifications.png