Cognilytica is now part of PMI!

A feature is any measurable input that can be used in a predictive model, similar to a dimension, and often used interchangeably with dimension. However, the term feature is often used in the context of supervised learning, especially neural network approaches as features might be learned and different at each layer of the neural network, whereas in other algorithmic approaches, the dimension or features are established, selected, or extracted prior to the model training.