Structured prediction is an umbrella term for machine learning and regression techniques that involve predicting structured objects. For example, the problem of translating a natural language sentence into a syntactic representation such as a parse tree can be seen as a structured prediction problem in which the structured output domain is the set of all possible parse trees. Structured prediction generalizes supervised learning where the output domain is usually a small or simple set.
Probabilistic graphical models form a large class of structured prediction models. In particular, Bayesian networks and random fields are popularly used to solve structured prediction problems in a wide variety of application domains including bioinformatics, natural language processing, speech recognition, and computer vision.
Similar to commonly used supervised learning techniques, structured prediction models are typically trained by means of observed data in which the true prediction value is used to adjust model parameters. Due to the complexity of the model and the interrelations of predicted variables the process of prediction using a trained model and of training itself is often computationally infeasible and approximate inference and learning methods are used.
Another commonly used term for structured prediction is structured output learning.
See also 
- Conditional random field
- Hidden Markov model
- Recurrent neural network, in particular Elman networks (SRNs)
- Noah Smith, Linguistic Structure Prediction, 2011.