In DSX Local using SPSS Modeler as a data preparation tool where it would be possible to use a custom algorithm in R for the final model.. Currently working on data preparation in SPSS Modeler and than building model in R are two separate processes. Tables produced at the end of the data preparation could be significant and therefore they the ability to use a Modeler flow as a source node for data scientists notebook would thereby keep all the data in memory/temp files rather than physically writing/importing as a data set. This would work a little bit like the Analytic Data View (ADV) in SPSS Collaboration & Deployment Services.
Why is it useful?
|Who would benefit from this IDEA?||Data scientists would have more smooth and efficient process of data prep and model building whilst combining different tools|
How should it work?
|Submitting Organization||F2F Sales|