Please enable JavaScript.

Coggle requires JavaScript to display documents.

Chapters 2-7 (Success, Risk and Continuation (Foresee risks (Model risks…

- - - - is management on board
        
        can model drivers be visualized
        
        how much value can the model produce
  - - - Risk of poor model/ insufficient data
    - - Risk of being ethically discriminatory
        
        ex: privacy issues
    - - the environment for your data changes drastically after a project
- - - - Makes subject matter expertise more valuable
        
        Makes subject matter expertise more valuable
        
        Not "automatic"
        
        ex: run by subject matter experts who understand Machine learning conceptually
        
        context-specific tools & general platforms
        
        context-specific: w/i system or for specific purpose
        
        general platforms: for general purpose ML
        
        1 more item...
- - - - Decide if problem is worth addressing and how to go about it
        
        How to improve bottom line
        
        Project statements must:
        
        Be in language of business
        
        Specify resulting actions
        
        How is bottom line affected
- - - - know where to collect data from
- - - - Classification
        
        Predicts category new case belongs
      - regression
        
        predict the target numeric values
- - - - Not alway obvious which unit you are using
        
        Lean on subject matter expert