Precision, Recall and F1-Score using R

Precision, Recall and F1-Score using R

In R, you can calculate Precision, Recall (also known as Sensitivity), and the F1-Score using the results from a confusion matrix. This is commonly done in the context of classification tasks. Here's how you can do it:

1. Create a Confusion Matrix

First, you need a confusion matrix, which you can generate using the table() function or specific functions from packages like caret.

Assume you have the actual and predicted classifications:

actual <- factor(c('yes', 'no', 'no', 'yes', 'yes', 'no', 'yes', 'no', 'no', 'yes')) predicted <- factor(c('yes', 'yes', 'no', 'no', 'yes', 'yes', 'yes', 'no', 'no', 'no')) conf_matrix <- table(Predicted = predicted, Actual = actual)

2. Calculate Precision, Recall, and F1-Score

Precision and Recall can be calculated as follows:

  • Precision (Positive Predictive Value): TP / (TP + FP)
  • Recall (Sensitivity, True Positive Rate): TP / (TP + FN)
  • F1-Score: 2 * (Precision * Recall) / (Precision + Recall)

Where TP is True Positives, FP is False Positives, and FN is False Negatives.

TP <- conf_matrix[2, 2] # Assuming 'yes' is the positive class FP <- conf_matrix[2, 1] FN <- conf_matrix[1, 2] precision <- TP / (TP + FP) recall <- TP / (TP + FN) f1_score <- 2 * (precision * recall) / (precision + recall)

3. Print the Results

cat("Precision:", precision, "\n") cat("Recall:", recall, "\n") cat("F1 Score:", f1_score, "\n")

Additional Notes:

  • Confusion Matrix: Make sure you understand the orientation of your confusion matrix. The arrangement of TP, FP, FN, and TN can vary depending on how the matrix is constructed.
  • Factor Levels: Ensure that the factor levels for your actual and predicted vectors are consistent and in the same order.
  • Libraries for Convenience: Although this example uses base R, there are libraries like caret, e1071, and Metrics that offer functions to calculate these metrics more conveniently and with additional functionality.

Example Using Caret Package:

If you're using the caret package, it becomes quite straightforward:

library(caret) conf_matrix <- confusionMatrix(predicted, actual) cat("Precision:", conf_matrix$byClass['Pos Pred Value'], "\n") cat("Recall:", conf_matrix$byClass['Sensitivity'], "\n") cat("F1 Score:", conf_matrix$byClass['F1'], "\n")

In this example, confusionMatrix() from the caret package calculates a variety of metrics, including Precision, Recall, and F1 Score.

More Tags

powershell-4.0 kotlin python-extensions kendo-chart pytz pygame-clock configparser nsarray webstorm docker-volume

More Programming Guides

Other Guides

More Programming Examples