📌 What is Accuracy in Machine Learning?

Accuracy is a basic metric used to evaluate classification models. It tells us how often the model is correct overall.

Formula:

Accuracy=Number of Correct PredictionsTotal Number of Predictions\text{Accuracy} = \frac{\text{Number of Correct Predictions}}{\text{Total Number of Predictions}}

Or in terms of confusion matrix terms:

Accuracy=TP+TNTP+TN+FP+FN\text{Accuracy} = \frac{TP + TN}{TP + TN + FP + FN}

BUT — accuracy can be misleading, especially if the dataset is imbalanced (e.g., 95% of one class and only 5% of another).

📌 What is a Confusion Matrix?

A confusion matrix is a table that helps visualize the performance of a classification model — particularly what it gets right and where it gets confused.

It breaks down predictions into categories to show:

• Correct predictions (True Positives & True Negatives)
• Mistakes (False Positives & False Negatives)

📊 Confusion Matrix Layout (for binary classification)

🧠 Understanding TP, FP, TN, FN:

✅ True Positive (TP)

• Model predicted Positive, and it was actually Positive.