Classification problems are one of the most common scenarios we face in data science. This course will help us understand and apply common algorithms to make predictions and drive decision-making in business. From Logistic Regression to KNN and SVM models, we'll learn how to implement techniques in Excel and Python and how to create loops to run models in parallel. Since model evaluation is so important, we'll dedicate a whole chapter to interpreting model outputs with evaluation metrics and the confusion matrix. With this, we'll learn about false negatives, and false positives, and consider the impacts these may have on specific business scenarios. Finally, we'll have a brief insight into more advanced classification techniques such as feature importance, SHAP values, and PDP plots.

This module introduces the fundamental concepts of classification in machine learning. It covers what classification is, the machine learning ecosystem surrounding it, and different types of classification problems (binary, multi-class, and multi-label). The module also explores common classification use cases, visualization techniques for classification problems, and provides an introduction to various classification algorithms.

This module provides a comprehensive introduction to logistic regression, a foundational classification algorithm. It explains how to visualize logistic regression, its underlying assumptions, and key concepts such as probability, odds, and log odds. The module demonstrates how to interpret logistic regression coefficients and odds ratios through practical examples. It also provides hands-on implementation in both Excel and Python with step-by-step guidance on building and analyzing logistic regression models.

This module explores various classification algorithms beyond logistic regression. It covers Naïve Bayes, K-Nearest Neighbors, Support Vector Machines, Decision Trees, and Random Forests. For each algorithm, the module explains core concepts and provides practical examples. The second half of the module demonstrates Python implementation of these classification techniques, including data importing, exploration, and building predictive models using different algorithms. The module provides a comprehensive overview of how to select and apply different classification approaches depending on the problem context.

This comprehensive module focuses on techniques for evaluating classification model performance. It begins with the confusion matrix and various evaluation metrics such as accuracy, precision, recall, and F-score. The module explores the trade-offs between precision and recall, discusses when accuracy may not be the best choice for evaluation, and explains ROC curves and AUC. It addresses common challenges like underfitting and overfitting. The module then transitions to Python implementation, demonstrating how to set up evaluation loops, calculate metrics, create confusion matrices, and interpret ROC curves. The final section covers interpretability versus explainability, feature importance, partial dependence plots, and SHAP values for understanding individual predictions.

This brief module summarizes the key concepts covered throughout the Classification course. It recaps the essential topics from classification fundamentals, logistic regression, other classification algorithms, and model evaluation techniques, providing a comprehensive conclusion to the course material.

This final assessment module tests the students' understanding of classification techniques covered in the course. Students apply their knowledge to solve classification problems, implement various algorithms, evaluate model performance, and interpret results in the context of real-world data and business scenarios.