Machine Learning with Business Rules on IBM Z: Acting on Your Insights
By Mike Johnson, Chris Backhouse, Stéphane Faure, David Griffiths, Yann Kindelberger, Ke Wei Wei & Hao Zhang
- Release Date: 2019-12-11
- Genre: Computers
This Redpaper introduces the integration between two IBM products that you might like to consider when implementing a modern agile solution on your Z systems. The document briefly introduces Operational Decision Manager on z/OS and Machine learning on z/OS. In the case of Machine Learning we focus on the aspect of real-time scoring models and how these can be used with Business Rules to give better decisions.
Note: Important changes since this document was written:
This document was written for an older release of Operational Decision Manager for z/OS (ODM for z/OS). ODM for z/OS 8.9.1 required the writing of custom Java code to access a Watson Machine Learning for z/OS Scoring Service (this can be seen in ). Since that time ODM for z/OS version 8.10.1 has been released and much improves the integration experience. Integrating the two products no longer requires custom Java code. Using ODM for z/OS 8.10.1 or later you can use an automated wizard in the ODM tooling to:
Browse and select a model from Watson Machine Learning
Import the Machine Learning data model into your rule project
Automatically generate a template rule that integrates a call to the Watson Machine Learning scoring service
Download and read this document for:
Individual introductions to ODM for z/OS and Machine learning
Discussions on the benefits of using the two technologies together
Information on integrating if you have not yet updated to ODM for z/OS 8.10.1
For information about the machine learning integration in ODM for z/OS 8.10.1 see "IBM Watson Machine Learning for z/OS integration" topic in the ODM for z/OS 8.10.x Knowledge Center at:
https://www.ibm.com/support/knowledgecenter/en/SSQP76_8.10.x/com.ibm.odm.zos.develop/topics/con_ml_integration_overview.html