easy5 pts · first answer countsclassification-metricsclass-imbalance
High accuracy on rare events
Only 1% of transactions are fraud. A model that predicts "not fraud" for every transaction scores 99% accuracy. What is the main problem?
Discussion 0
Join free to post in the discussion.
Related questions
- Precision from a confusion matrixData Scientist
- Metric when missing a case is costlyData Scientist
- What an AUC of 0.8 meansData Scientist
- F1 from precision and recallData Scientist
- Which prediction hurts log loss mostData Scientist
- Using a model on rare positivesData Scientist
- Brier score for two predictionsData Scientist
- AUC after squaring the scoresData Scientist