Model accuracy measures how often the classifier's predictions match the true labels, expressed as a percentage. A 90% accuracy rate means 9 out of 10 classifications are correct. Accuracy is a useful shorthand but can hide poor performance on minority categories. For imbalanced datasets (where some reasons are rare), precision, recall, and F1 scores provide better evaluation.
← Glossary
Model Accuracy
The percentage of classifications the AI model gets correct across all categories.
Related terms
- AI Classifier — A machine learning model that automatically categorizes inputs—like return reasons or customer feedback—into predefined groups.
- Training Data — The labeled examples used to teach an AI model how to categorize returns correctly.
- False Positive — When the AI incorrectly assigns a return reason that doesn't match the actual reason.
- False Negative — When the AI fails to identify the correct return reason and assigns an incorrect one.