Applied Machine Learning and Visualization

This unit advances students’ practical Python data science skills by covering supervised and unsupervised learning algorithms, feature selection, hyperparameter tuning, ensemble methods, and strategies to prevent overfitting. It also teaches applied data engineering and visualization techniques—including database access, date handling, geospatial plotting, environment setup, hardware acceleration, parallel processing, and preparing datasets and example code—to bridge exploratory analysis and production-ready workflows.

Learning Objectives

• Apply clustering and classification algorithms (DBSCAN, k‑nearest neighbors, support vector machines, Naïve Bayes, and boosting) in Python to build predictive models and compare their performance using appropriate evaluation metrics.
• Analyze multivariate datasets to perform feature selection, hyperparameter tuning, and overfitting mitigation so models generalize effectively to new data.
• Demonstrate data engineering and visualization workflows by loading and managing data from relational databases, handling date fields, producing labeled and geospatial plots, and conducting exploratory data analysis with the Python ecosystem.
• Evaluate and optimize computational performance by configuring Anaconda environments, using hardware acceleration and parallel processing, and implementing chunked processing strategies for large datasets.