Part Four in a series of four on the Intelligent Enterprise Machine Leaning Framework. Read Part One, Part Two, Part Three.
The details of the machine learning (ML) framework presented in the previous blog can be used as a reference for the tasks which you should consider when you are delivering an ML project for the Intelligent Enterprise. Remember that every project has different requirements, so these tasks, and their sequence, are recommendations that you will probably have to adjust accordingly.
The material presented in this blog is a simplified summary that you can use. Here is the summary information for each phase, showing the tasks and the potential deliverables:
In the Define phase of this framework, to gain a clear understanding of the requirements for an ML project, I suggest you try to work with your customer and use a design thinking approach. However, in some situations an iterative approach with regular meetings with the customer might be more feasible. You should therefore consider design thinking as a potentially useful approach, but it isn't mandatory.
This ML framework contains a lot of detail to guide you through the ML project delivery process for the Intelligent Enterprise. It can be used by those new to delivering ML projects, citizen data scientists, or by seasoned data science experts. Importantly, it can be used by anybody who currently uses CRISP-DM.
However, I understand that some seasoned practitioners might need to make minor changes based on budget constraints or project deadlines, but they can still use this framework as a guide.
In future developments, I hope to refine this to make a distinction between different project end states---POCs, pilots, prototypes, production deployments.
For an in-depth look into the intelligent possibilities for your business, review the August 2018 Forrester Consulting study, Powering The Intelligent Enterprise With AI, Machine Learning, And Predictive Analytics, commissioned by SAP.
DMAIC was first developed as a data science framework by colleagues at KXEN, who were responsible for developing the automated analytics engine we now refer to as SAP Predictive Analytics.
Colleagues at SAP have helped me shape this framework and given me valuable feedback so that it supports SAP technology and processes.
The details of the machine learning (ML) framework presented in the previous blog can be used as a reference for the tasks which you should consider when you are delivering an ML project for the Intelligent Enterprise. Remember that every project has different requirements, so these tasks, and their sequence, are recommendations that you will probably have to adjust accordingly.
The material presented in this blog is a simplified summary that you can use. Here is the summary information for each phase, showing the tasks and the potential deliverables:
Summary
In the Define phase of this framework, to gain a clear understanding of the requirements for an ML project, I suggest you try to work with your customer and use a design thinking approach. However, in some situations an iterative approach with regular meetings with the customer might be more feasible. You should therefore consider design thinking as a potentially useful approach, but it isn't mandatory.
This ML framework contains a lot of detail to guide you through the ML project delivery process for the Intelligent Enterprise. It can be used by those new to delivering ML projects, citizen data scientists, or by seasoned data science experts. Importantly, it can be used by anybody who currently uses CRISP-DM.
However, I understand that some seasoned practitioners might need to make minor changes based on budget constraints or project deadlines, but they can still use this framework as a guide.
In future developments, I hope to refine this to make a distinction between different project end states---POCs, pilots, prototypes, production deployments.
Learn More
For an in-depth look into the intelligent possibilities for your business, review the August 2018 Forrester Consulting study, Powering The Intelligent Enterprise With AI, Machine Learning, And Predictive Analytics, commissioned by SAP.
Acknowledgments
DMAIC was first developed as a data science framework by colleagues at KXEN, who were responsible for developing the automated analytics engine we now refer to as SAP Predictive Analytics.
Colleagues at SAP have helped me shape this framework and given me valuable feedback so that it supports SAP technology and processes.
- SAP Managed Tags:
