Denetimli Öğrenme

It includes classification, where targets are categories, and regression, where targets are numerical quantities. The quality and meaning of the target labels are central to the result.

Key takeaways

• Define labels before collecting them.
• Keep related records from leaking across evaluation splits.
• Measure the mistakes that matter to the workflow.

Each training example tells the algorithm what output is desired for an input. A loss function converts prediction errors into a quantity the training procedure can optimize. The choice of loss shapes learning; the metric used to judge the final workflow may be different. Labels can come from measurements, later outcomes, or annotation. Examine disagreements and ambiguous cases rather than assuming every recorded answer is correct. If the label captures an old decision process, the model can reproduce that process’s limitations. Split the data to match how the model will encounter new cases. Random row splits can leak information when repeated records describe the same subject. Forecasts generally need time-respecting evaluation. Fit preprocessing steps only on the training partition before applying them to validation and test examples. After training, inspect performance for relevant classes and operating conditions. Class imbalance can make overall accuracy misleading. Decide how uncertain or unfamiliar inputs should be handled, and retain a route for correcting labels and reviewing systematic mistakes.

1. In a constructed test with 40 urgent messages, a classifier catches 30 and misses 10. It also flags 20 ordinary messages.
2. Urgent-message recall is 30/40 = 75%. Precision among flagged messages is 30/(30+20) = 60%.
3. Ask whether reviewing 50 flagged messages to find 30 urgent ones is useful for the team’s capacity and priorities.

The arithmetic describes a hypothetical workload, not a reported product benchmark.