Type Package Title Node Harvest for regression and classification Version 0.6 Date 2013-11-28

Imports graphics, quadprog,randomForest Description Node harvest is a simple interpretable tree-like estimator for high-dimensional regres-sion and classification. A few nodes are selected from an initially large ensem-ble of nodes, each associated with a positive weight. New observations can fall into one or sev-eral nodes and predictions are the weighted average response across all these groups. The pack-age offers visualization of the estimator. Predictions can return the nodes a new observa-tion fell into, along with the mean response of training observations in each node, offering a sim-ple explanation of the prediction