1
Introduction
2
Preface
2.1
The original story
2.2
What it does
3
Our data
3.1
Data: FOIA Requests
3.2
Success metrics (editorial choice)
3.3
Features (editorial choice)
4
Analysis
4.1
Designing our model
4.1.1
Classification problems
4.2
Selecting an algorithm
4.3
Training our algorithm
4.4
Evaluation metrics
4.4.1
Accuracy
4.4.2
Dummy classifier
4.4.3
Confusion matrix
4.4.4
Explainability
5
Other classifiers
5.1
Logistic Regression
5.2
Decision Trees
5.3
Random Forest
6
Feature selection and engineering
6.1
Leaving out our best feature
6.1.1
Setting up our features
6.1.2
K-nearest neighbors
6.1.3
Logistic Regression
6.1.4
Decision tree
6.1.5
Random forest
6.1.6
Summary
7
Explaining predictions
8
Percent probability
9
Review
10
Discussion topics
3
Our data