Statistical Machine Learning
Statistical Machine Learning
Learning environment taught by Prof. Dr. Jodavid Ferreira. Centralized access to interactive slides, practical exercise lists, code repositories, and the course project.
đź“š Course Topics and Lectures
Explore the course content organized into teaching modules. Click the button on each lecture to open the interactive full-screen presentation.
Course Presentation
Artificial Intelligence Concepts
Introduction to R
Introduction to Python
Data and Models
R code - Example of Feature Selection
Prediction vs Inference
Assessing model accuracy
R code - Example of Ordinary Least Squares (OLS)
R code - Example of Kappa Coefficient - pt_BR
R code - Example of Kappa coefficient Implementation - pt_BR
Resampling Methods
Explainability in ML models - LIME, SHAP, and DALEX
Algorithmic Fairness and Bias Mitigation in Machine Learning
Parametric Methods for Regression
Nonparametric Methods for Regression
Nonparametric Methods for Regression Tree Based
Classification Methods
Clustering Methods
đź“ť Exercise Lists
Practical exercise lists to consolidate knowledge and prepare for the course’s theoretical and practical assessments.
1st Assessment
Lists 01 to 082nd Assessment
Lists 09 to 13Link to Codes
đź’» Code & Datasets
Access the source code demonstrated in class and the datasets used for hands-on experimentation.
Lecture Code
Official Repository: GitHub Link
Example Dataset
Kaggle DatasetDataset used in hands-on activities and demonstrations of exploratory analysis and predictive modeling on federal highways.
đź“‘ Course Project & Activities
Applied Project in Statistical Machine Learning
The course project covers the full cycle of solving a real-world data problem: problem definition, data collection and cleaning, in-depth exploratory analysis with data visualization, predictive modeling with the algorithms studied, and an executive presentation of the insights.
Problem definition and selection of a relevant dataset.
Handling missing values and outliers, feature engineering, and visualizations.
Training and comparing Statistical Machine Learning models.
Final pitch with storytelling and a clean code repository.
đź’ˇ Instructions for Using the Notebooks in Google Colab
How to Run the Scripts in Google Colab
Follow these steps to run, edit, and save your code directly in your personal or institutional Google Drive.
Accessing the Script
The notebook is shared in view-only mode. To preserve the integrity of the original code, the base file cannot be edited directly.
Save a Copy
In the top menu of Google Colab, click File → Save a copy in Drive. An independent copy will be saved to your Google account.
Run & Customize
Open your copy and feel free to run each cell, try new parameters, upload new files, and build your own solutions.