Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapters 2-7 Automated Machine Learning (Automated Machine Learning…
Chapters 2-7 Automated Machine Learning
Automated Machine Learning
Tools used
Context Specific
Implemented in another system
general platforms
Open Source
Python
R
Commercial
Provided by outside company usually for a price
Criteria
Accuracy
Productivity
Ease of Use
Understanding and Learning
Resource availability
Process Transparency
Generalizable
Recommended actions
Steps
Specify Business Problem
important to start here to show why we are working
Problem Statements
Acquire subject matter expertise
important to fully understand the field in which you are working
Decide on Unit of Analysis
The what, where, when, and who of your analysis
Defines exactly what you are analyzing and why you are analyzing it.
Define Prediction Target
Choosing what you want to be looking for at the end of your analysis
Assumptions and associations cannot be made without a target
Identify Success, Risk, and Continuation
Success
What criteria make this model useful and helpful to the people using it
What will this model if it is successful give to the people using it
Risks
Looking into risks that may make it difficult to continue using the model
Ethical, Practical, Logistical, Cost, Time
Continuation
consider risks and successes of model to decide if it is useful