Practical Exercise. Analysis of Metrics for Optimizing Supervised Regression and Classification Models

Author: Gregorio Mendoza Serrano

In this practical exercise, we will study how to use the metrics previously studied in theory to solve practical problems.

The competencies associated with these teaching units are as follows:

• Students should have the ability to gather and interpret relevant data to make judgments that include reflection in the field of machine learning.

• Design an experimental framework considering the most appropriate methods for data capture, processing, storage, analysis, and visualization.

• Combine mathematical, statistical, and programming fundamentals to develop solutions to problems in the field of data science.

The objectives we aim to achieve with this practical notebook are:

• Gain a clear understanding of the metrics that measure the performance of supervised machine learning models for regression and classification.

• Work on related points in machine learning, such as cross-validation, overfitting/underfitting issues, and data balancing problems.

• Work with, understand, evaluate, compare, and interpret the main metrics in different machine learning algorithms for regression and classification applied to real-world problems.