Exploratory Analysis and Modeling

This unit advances foundational Python skills into applied data science workflows: validating and preparing data, creating informative visualizations, engineering features, and applying unsupervised and supervised modeling techniques. Students will also establish secure, reproducible environments and pipelines while measuring and improving execution performance to prepare for production and advanced modeling tasks in subsequent units.

Learning Objectives

• Analyze raw datasets to detect and correct data quality issues using validation techniques, categorical handling, array operations, and unstructured file I/O
• Apply feature engineering and visualization techniques (axis/tick/grid control, scatterplots, time series, and line styling) to explore relationships and prepare inputs for modeling
• Demonstrate dimensionality reduction and unsupervised learning by performing PCA/factor analysis and hierarchical clustering, and interpret results to inform feature selection
• Evaluate and implement supervised model workflows—including logistic regression, regularization, cross-validation, and tree-based models—while building secure, reproducible pipelines, measuring timing/performance, and configuring environments (e.g., Anaconda)