Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter 7: Decision Analytic Thinking (Predictors (Decision Tree (One…
Chapter 7: Decision Analytic Thinking
Classification
Easy to measure
Too simplistic
Classifier Accuracy
Accuracy Equation
Confusion Matrix
Separates out decisions
Shows how one class is being confused for another
False positives and false negatives
Class Confusion
Makes Errors Equally Important
Cancer Example
False positive: told you have cancer when you don't
False negative: told you don't have cancer when you do
Expected value
Framework for organizing thinking
Structure for the problem
Elements of the analysis that can be extracted from data
Elements of analysis that need to be acquired from other sources
Individual Decisions
Collections of Decisions
Aggregate: what is the expected value
Where do probability outcomes come from?
Costs/Benefits
Cost-benefit matrix
Cannot be estimated by data
Classify
False Positive
False negative
True Positive
Offered/Buys
True Negative
Not offered/not bought
Predictors
Persistance
Climatology
Better than random guessing
Majority Classifier
Decision Tree
One internal node
The Root Node
Decision Stump