Fスコア、別名 F1スコア, is an 評価指標です for 二値分類 models that balances precision and recall. It is particularly useful when the class distribution is imbalanced, meaning one class significantly outnumbers the other. The F-Score provides a single metric that combines both the precision (the ratio of true positive predictions to the total predicted positives) and recall (the ratio of true positive predictions to the actual positives).
Fスコアの計算式は次の通りです：
F1スコア = 2 * (精度 * 再現率) / (精度 + 再現率)
ここで：
- 精度: The number of true positives divided by the number of true positives plus false positives.
- 想起: The number of true positives divided by the number of true positives plus false negatives.
The F-Score ranges from 0 to 1, where a score of 1 indicates perfect precision and recall, while a score of 0 indicates the worst performance. An advantage of the F-Score is that it gives equal weight to precision and recall, making it more informative than accuracy 特に一方のクラスがもう一方より重要な場合に役立ちます。
実際には、Fスコアはさまざまな分野で広く使用されています。 自然言語処理, medical diagnosis, and any domain where the cost of false positives and false negatives is significant. By optimizing for the F-Score, practitioners can ensure a more balanced approach to model evaluation and improve overall model performance.