Accuracy

Definition

• Accuracy Score is the proportion of correct predictions out of all predictions made. In other words, it is the percentage of correct predictions.
• Accuracy can have issues with Imbalanced Datasetswhere there is more of one class than another.

Formula

• The formula for accuracy is: $\text{Accuracy}=\frac{TN+TP}{\text{Total}}$ In the context of Classification problems, particularly binary classification, TN and TP are components of the confusion matrix:

• TP (True Positive): The number of instances that are correctly predicted as the positive class. For example, if the model predicts a positive outcome and it is indeed positive, it counts as a true positive.

• TN (True Negative): The number of instances that are correctly predicted as the negative class. For example, if the model predicts a negative outcome and it is indeed negative, it counts as a true negative.

The Confusion Matrix also includes:

• FP (False Positive): The number of instances that are incorrectly predicted as the positive class. This is also known as a “Type I error.”
• FN (False Negative): The number of instances that are incorrectly predicted as the negative class. This is also known as a “Type II error.”

These metrics are used to evaluate the performance of a classification model, providing insights into not just accuracy but also precision, recall, and other performance measures.

Exploring Accuracy in Python

To explore accuracy in Python, you can use libraries such as scikit-learn, which provides the accuracy_score function. This function compares the predicted labels with the true labels and calculates the accuracy.

Example Usage

from sklearn.metrics import accuracy_score
# Assuming pred and y_test are defined
accuracy = accuracy_score(y_test, pred)
print("Prediction accuracy: {:.2f}%".format(accuracy  100.0))

• Make sure to replace pred and y_test with your actual prediction and test data variables.