Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter: 2-7 (Chapter 2. Automating Machine Learning (2.4 Eight Criteria…
Chapter: 2-7
Chapter 2. Automating Machine Learning
Machine Learning Life Cycle
Define Project Objectives
Acquire & Explore Data
Model Data
Interpret & Communicate
Implement, Document & Maintain
2.1 What is Automated Machine Learning
"off-the-shelf methods"
Increase efficiency
2.2 What Automated Machine Learning is Not
Not automatic ML
Which ideas are worth solving
frees data scientist from manually testing
2.3 Available Tools and Platforms
Context-Specific Tools and General Platforms
Salesforce Einstein
Wise,io
Open Source
Python R
Commercial Vendor
General Purpose Machine Learning
2.4 Eight Criteria for AutoML Excellence
Accuracy
Productivity
Ease of Use
Understanding & Learning
Resource Availability
Process Transparency
Generalizable across contexts
Recommend actions
Chapter 7 Success, Risk, and Continuation
7.1 Identify Success Criteria
Who will use the model
Is the managment on board with the project
Can the model drivers be visualized
How much can the model produce
7.2 Forsee Risks
Creativity
Devils Advocate
Model Risks
Not allowed to access applicants loans
Ethical Risks
Ethical Risks
Privacy
Cultural Risks
Marketing and HR
Evironmental Risks
Black Swan Phenomenon
7.3 Decide Whether to Continue
Succeed or fail quickly to minimize risk
7.4 Excercise
Chapter 5 Decide on Unit of Analysis
5.1 What is a Unit of Analysis?
What, Who, Where, When
Can have multiple outcomes
HR ex.
5.2 How to Determine Unit of Analyis
Think about what the rediction target is
Lending Club
Learn Subject Matter
Learn to share knowledge of problem
5.3 Excercise
Chapter 6 Define Prediction Target
6.1 What is a Prediction Target
Hint at the target without specifying it
Future
6.2 How is the Target Important for Machine Learning?
NO way for humans or machines to learn
Classification
Predicts category to which case belongs
Regression
Data set to prediction
6.3 Excercise/Discussion
Chapter 3 Specify Business Problems
3.1 Why Start with a Business Problem?
Which customers are likely to buy
Why customers do not purchase
Why customers purchase
Why customers are dissatisfied
Why customers do no renew
What leads to extensive
Which Internet users
Which pages would benefit
3.2 Problem Statement
Is it presented in project statement
Is it specific
How could it impact the bottom line
Chapter 4 Acquire Subject Matter Expertise
4.1 IMportance of Subject Matter Expertise
Important for detecting potential obstacles
Data collection
4.2 Excercise
Hospital Ex
Google/Google Scholar