Appropriate evaluation metrics are used based on the problem type (classification vs. regression), to assess how well the model predicts.
The aim is to improve accuracy but also to generalize and avoid biases and Overfitting.
- Performance Assessment: Models are evaluated on a testing set using metrics relevant to the problem type.
- Generalization and Bias: Evaluation includes assessing how well the model generalizes to new data and identifying any biases.
For categorical classifiers: Classification Metrics: Use metrics such as accuracy, precision, recall, F1-score, and confusion matrix to evaluate performance.
For regression tasks: Regression metrics: Metrics like Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and R-squared (R²) are used.
Cross Validation is a technique used to assess the performance of a model by splitting the data into multiple subsets for training and testing to assesses performance and generalization. It helps detect Overfitting, provides reliable performance estimates.