Training data consists of historical return records where the correct category is already known. For example, returns manually tagged by support staff with reasons like 'defective' or 'wrong size.' The model learns patterns from these examples to predict reasons for new returns. Higher quality and quantity of training data produces more accurate classifiers.
← Glossary
Training Data
The labeled examples used to teach an AI model how to categorize returns correctly.
Related terms
- Machine Learning Model — A mathematical system trained on historical data to make predictions or classifications on new data.
- Labeled Data — Training examples where humans have already assigned the correct category or answer.
- Model Accuracy — The percentage of classifications the AI model gets correct across all categories.
- Feature Engineering — The process of selecting and transforming raw data into the most useful inputs for an AI model.