Supervised Learning

Context

• With Supervised Learning, the objective is to find the best probabilistic characterization of a Target Node, i.e, producing a useful predictive model.
• This differs from Unsupervised Structural Learning, which attempts to find the best representation of the Joint Probability Distribution sampled by observations (or particles) recorded in a dataset.
• In early editions of BayesiaLab, Supervised Learning was also known as Characterization of the Target Node.

Supervised Learning Algorithms Available in BayesiaLab

• Augmented Markov Blanket
• Augmented Naive Bayes
• Markov Blanket
• Minimal Augmented Markov Blanket
• Naive Bayes
• Semi-Supervised Learning
• Sons and Spouses
• Tree Augmented Naive Bayes