Predictive Analytics and Data Mining

Decision modeling is used for framing analytic project requirements.

  • Organizations generally use standard requirements approaches for analytic projects. They define the data that needs to be provided, they identify the analytic technology to be used and they define the workflow for delivering this data to decision-makers.
  • A more effective way to define requirements for analytic projects, to frame those requirements accurately, is to model the decision-making to be improved. This Decisions First requirements approach provides critical information for successful analytic projects, complements workflow requirements, and correctly identifies the data that will be required for the effort.

In DecisionsFirst Modeler, a predictive analytics or data mining result is captured as a Knowledge Source. The link between a decision and a Knowledge Source are authority requirements (dashed line).

 A first level of refinement, you can identify authority requirements for a Knowledge Source, for example the data you would look at to derive the predictive analytics or data mining result. For some analytic projects where more detail about the actual analytic project is needed, you would create an analytic effort object and link it to the relevant Knowledge Source. 

Here is a short video on creating an analytic effort object.
Also in this video, starting at about 24',  James describes decision requirements, 
Here you will see the high level decision with some analytic KS, and then the decision decomposition, etc.

To learn more,download our white paper, Framing Analytic Requirements with Decision Modeling.