Unit3_Evaluating_Models_Visuals_ClassX.pptx for grade x

evaluating models using examples

1. Unit 3: Evaluating Models Artificial Intelligence – Class X Detailed with Visuals

2. Importance of Model Evaluation - Evaluation checks how well the AI model performs - Ensures the model is reliable and accurate - Helps improve performance by identifying weaknesses - Prevents wrong or misleading predictions

3. Splitting the Training Data - Data is divided into Training and Testing sets - Training Data: Used to teach the model - Testing Data: Used to evaluate performance - Example: 80% training, 20% testing

4. Accuracy & Error Example - Accuracy = (Correct ÷ Total) × 100 - Error = (Wrong ÷ Total) × 100 - Example: 90 correct, 10 wrong → Accuracy = 90%, Error = 10%

5. Confusion Matrix Example - Confusion Matrix compares actual vs predicted results - TP = correctly predicted positives - TN = correctly predicted negatives - FP = wrongly predicted positives - FN = missed positives Predicted: Positive Predicted: Negative Actual: Positive TP (True Positive) FN (False Negative) Actual: Negative FP (False Positive) TN (True Negative)

6. Precision - Precision = TP ÷ (TP + FP) - Out of predicted positives, how many were correct? - High precision = fewer false alarms

7. Recall - Recall = TP ÷ (TP + FN) - Out of actual positives, how many were found? - High recall = fewer missed cases

8. F1 Score - F1 Score = 2 × (Precision × Recall) ÷ (Precision + Recall) - Balances precision and recall - Useful for fair evaluation

9. Ethical Concerns - Bias → unfair results - Transparency → knowing how AI makes decisions - Accuracy → models must be reliable - Ensures fairness & trust

10. Summary - Evaluation measures AI model performance - Train-Test split helps testing - Accuracy & Error = basic measures - Confusion Matrix, Precision, Recall, F1 = detailed metrics - Ethics are important for fairness