Software

Version 4.5 of the DTREG predictive modeling program has been released. DTREG builds models using Support Vector Machines (SVM), Decision Trees and Logistic Regression. Decision tree models include both classical, single-tree models and the more accurate TreeBoost models consisting of a series of boosted trees. DTREG also builds Decision Tree Forests consisting of many trees that "vote" on the result. You can build each type of model by simply selecting a menu item.

DTREG performs both classification and regression analysis, and it is capable of handling problems with hundreds of thousands of rows and thousands of variables. DTREG supports both categorical and continuous variables. Categorical variables can have either numeric or symbolic values (e.g., "Male"/"Female"). A Data Transformation Language (DTL) is built into DTREG for transforming variables, creating new variables and performing record selection.

A free demonstration copy of DTREG is available from http://www.dtreg.com

DTREG price range: $900 to $3000