Modeling Features and Functions

• Mines data in database where it resides within data modeling functionality


• Use IBM DB2 Enterprise decision trees, regression, association, and demographic clustering techniques


• Use Oracle 10g naïve Bayes and adaptive Bayes networks and SVM


• Uses predictive and classification techniques


• Browses the importance of the predictors decision trees and rule induction techniques, including CHAID, exhaustive CHAID, QUEST, and C&RT


• Browses and interactively creates splits in decision trees


• Browses, collapses, and expands decision rules


• Uses linear regression, logistic regression, and multinomial logistic regression


• Views model equations and advanced statistical output


• Uses clustering and segmentation techniques Kohonen networks, K-means, and TwoStep


• Chooses from detection algorithms including GRI, Apriori, sequence, and CARMA


• Filters, sorts, and creates subsets of association models


• Employs data reduction techniques factor analysis and principal components analysis


• Combines multiple models, or one model can be used to build a second model


• Creates multi-variable regression predictive models


• Attribute-based predictive input metrics


• Imports third-party predictive models using the PMML industry standard


• Metric-based predictive input metrics


• Conditional predictive input metrics


• Drill down capabilities