IBM SPSS Modeler v16.0 What's new
Here is a review of what is new in SPSS v16. Although it is not officially released yet, a document that explains what’s new has been linked to the internet. You can find it by clicking here. Here is our summary of the features.
Some of the new key features that caught my attention are:
- TM1 Source and Export nodes
- With the source node you will be able to use cube views and import data into Modeler
- With the export node you will be able to score data to an existing TM1 cube
- New R nodes
- In Modeler 15 FP2 an R node was introduced for modelling, R Model Build/Apply. The new release has two additional nodes, R Process and R Output. With the R Process node, you can use R scripts to do transformations and with the R Output node, you can use R scripts to perform data analysis and produce text and graphical outputs. Also R nodes will be installed as part of the base Modeler installation and not separately.
- R in database
- R nodes can be pushed back to some databases offering better performance. Oracle, Netezza and SAP HANA.
- Monte Carlo Simulation
- Simulation Source node – generate synthetic data, using a wide selection of statistical distributions
- Fitting node – can create a preconfigured source node
- Simulation Evaluation node – terminal node designed to evaluate fields resulting from a simulated analysis stream
- Python scripting – use python to automates processes. Legacy scripting language will continue to be supported.
- New Distinct node – Will handle duplicates records more effectively.
- New Receiver Operator Characteristic (ROC) Evaluation node chart type
- Area Under the Curve (AUC) and Gini metrics in Analysis node