Title :
Model-based classification trees
Author :
Geman, Donald ; Jedynak, Bruno
Author_Institution :
Dept. of Math. & Stat., Massachusetts Univ., Amherst, MA, USA
fDate :
3/1/2001 12:00:00 AM
Abstract :
The construction of classification trees is nearly always top-down, locally optimal, and data-driven. Such recursive designs are often globally inefficient, for instance, in terms of the mean depth necessary to reach a given classification rate. We consider statistical models for which exact global optimization is feasible, and thereby demonstrate that recursive and global procedures may result in very different tree graphs and overall performance
Keywords :
optimisation; statistical analysis; trees (mathematics); classification rate; decision trees; exact global optimization; mean depth; model-based classification trees; nonparametric estimation; performance; recursive designs; recursive procedure; tree graphs; Bayesian methods; Classification tree analysis; Dynamic programming; Entropy; Machine learning; Military computing; Pattern recognition; Statistical distributions; Testing; Tree graphs;
Journal_Title :
Information Theory, IEEE Transactions on
