Data Analysis

Machine Learning for Predictive Analytics

New features include:

• READ_TREE function to read the contents of each individual tree within the random tree forest.
• RF_PREDICTOR_IMPORTANCE function to measure the importance of predictors using the Mean Decrease Impurity (MDI) approach.
• Improvements to SVM that allow you to put weight on samples based on their classes, and to output raw scores.
• The ability to name one hot encoded columns with something meaningful, rather than integers for APPLY_ONE_HOT_ENCODER.

For more information, see Machine Learning for Predictive Analytics.