Exploratory Data Analysis and Modeling

This unit develops practical skills for cleaning, exploring, visualizing, and modeling data using Python, while introducing reproducible, secure environments and performance considerations. Students will expand on basic programming foundations to perform feature engineering, apply dimensionality reduction and clustering, build and evaluate simple predictive models, and optimize workflows for speed and reproducibility.

Learning Objectives

• Analyze structured and unstructured datasets to validate data quality, identify missing or inconsistent values, and apply appropriate cleaning and transformation steps using Python libraries (e.g., pandas, numpy).
• Apply feature engineering and transformation techniques—including encoding categorical variables, scaling, PCA/factor analysis, and custom transformations—and create advanced visualizations (scatterplots, time series, axis/tick/grid control) to support exploratory analysis.
• Demonstrate implementation of supervised and unsupervised methods (logistic regression, regularized linear models, hierarchical clustering, random forests) and use cross-validation and timing measures to assess model performance.
• Evaluate and document reproducible, secure data science workflows by configuring environments (Anaconda), constructing data pipelines, exporting data formats, and optimizing code for execution speed and scalability.