Data Science Cheat Sheet

Confusion Matrix and F1 Score

Confusion Matrix

	Predicted Positive	Predicted Negative
Actual Positive	True Positive (TP)	False Negative (FN)
Actual Negative	False Positive (FP)	True Negative (TN)

F1 Score = 2 * (Precision * Recall) / (Precision + Recall)

• 2 * (Positive Predictive Value * True Positive Rate) / (Positive Predictive Value + True Positive Rate)
• 2 * (TP) / (TP + FP + FN)

Key Evaluation Metrics

True Positive Rate (TPR), Sensitivity, Recall

• TPR = Sensitivity = Recall = TP / (TP + FN)

Precision (Positive Predictive Value)

• Precision = TP / (TP + FP)

Specificity (True Negative Rate, TNR)

• Specificity = TNR = TN / (TN + FP)

False Positive Rate (FPR)

• FPR = FP / (FP + TN)

Negative Predictive Value (NPV)

• NPV = TN / (TN + FN)

Accuracy

• Accuracy = (TP + TN) / (TP + TN + FP + FN)

Relationships between Key Concepts

TPR (Recall) and Precision:

• TPR represents the proportion of actual positives correctly predicted by the model, while Precision shows the proportion of predicted positives that are actually positive.
  • Increasing TPR (Recall) can sometimes reduce Precision, and vice versa.

FPR and Specificity:

• Specificity = (1 - FPR). In an ROC curve, FPR is plotted on the x-axis and TPR on the y-axis to visualize model performance.

F1 Score:

• Defined as the harmonic mean of Precision and Recall, emphasizing their balance. F1 Score = 2 * (Precision * Recall) / (Precision + Recall)

Accuracy:

• Accuracy reflects the overall model performance but may not be suitable in cases of class imbalance.