Applied Machine Learning and Data Engineering

This unit advances foundational Python skills into practical data science workflows by covering supervised and unsupervised modeling, feature selection, model tuning, and strategies for scaling and optimizing data pipelines. Students will also learn applied exploratory visualization, data ingestion from multiple sources (including relational databases and date/time handling), and use of Python tooling and hardware acceleration to prepare models and analyses for production-level tasks.

Learning Objectives

• Analyze and compare supervised and unsupervised models (DBSCAN, KNN, Naïve Bayes, SVM, boosting) by measuring performance, diagnosing overfitting, and applying cross-validation and hyperparameter tuning techniques
• Apply data engineering techniques in Python to ingest and preprocess data from multiple sources (CSV, relational databases), handle dates and annotations, and process large datasets in chunks or via parallel/hardware-accelerated execution
• Demonstrate feature selection and model-improvement strategies, including multivariate approaches, variable selection, and lazy/nearest-neighbor methods to improve predictive accuracy and computational efficiency
• Evaluate and create applied visualizations for exploratory data analysis and communication, including labeled plots, legends, annotations, and geospatial visualizations, using the Python data science ecosystem