| CPC G06N 20/20 (2019.01) [G06Q 20/4016 (2013.01)] | 19 Claims |
|
1. A computer-implemented method for providing transformations for an ensemble machine learning model comprising:
providing all of a plurality of base machine learning (ML) models of the ensemble machine learning (ML) model;
identifying all of a plurality of Derived Fields in all the plurality of base ML models in the ensemble ML model;
performing a Derived Field run prediction analysis for all of the plurality of Derived Fields;
computing the Derived Field Importance Weight for Field (DFIW4F) for all of the plurality of Derived Fields;
computing the Derived Field Importance Weight for Model (DFIW4M) for all of the plurality of Derived Fields;
clustering all the Derived Fields into a plurality of Derived Field clusters, wherein each Derived Field cluster is based upon the DFIW4M and the DFIW4F for the Derived Field;
sorting all the Derived Field clusters by best cluster based upon DFIW4M and DFIW4F;
running the base ML models based upon the Derived Fields in the best Derived Field cluster until sufficient base ML models have been run to output a final prediction of the ensemble ML model wherein the ensemble ML model is optimized to provide a faster final prediction; and
performing an update comprising:
updating the run time for all of the plurality of Derived Fields;
calculating an updated DFIW4M and an updated DFIW4F for all the plurality of Derived Fields;
clustering all the Derived Fields into a plurality of updated Derived Field clusters based upon the updated DFIW4M and the updated DFIW4F for all the plurality of Derived Fields;
sorting the updated Derived Field clusters by best cluster based upon the updated DFIW4M and DFIW4F; and
running the base ML models based upon the updated Derived Fields in the best Derived Field cluster until sufficient base ML models have been run to output an updated final prediction of the ensemble ML model.
|