Class Imbalance

Big problem for classifiers

Happens when an overwhelming majority of your data is a single class

Examples of class imbalance:

• Fraud - Most transactions aren’t fraudulent
• Disease - Most people won’t have the disease
• Product - Most people won’t buy any given product

It’s very common to have a rare positive signal surrounded by negatives

Options for dealing with class imbalance:

• Use class_weight parameter in sklearn
  • Penalizes the class with too many examples, helps the class with too few examples
• Create multiple synthetic examples of your negative features with slight differences
• Equalize your training set todo what does this mean? remove the extra rows?