Applied Functional Data and ML

This capstone unit integrates functional programming techniques in Haskell and Python to process binary and text data, build data pipelines, and develop applied machine learning solutions. Students will combine parsing libraries, data tooling, and ML frameworks while practicing reproducible coding habits and considering ethical and domain-specific (e.g., healthcare) implications.

Learning Objectives

• Analyze and compare methods for handling binary and text data in Haskell and Python by implementing and testing parsers and serializers using libraries such as bytestring, text, attoparsec, and aeson.
• Apply functional tooling and data-pipeline libraries (toolz, funcy, moo, stringsearch) along with SQLAlchemy and PrettyTable to ingest, transform, and present datasets within reproducible Jupyter notebook workflows.
• Demonstrate the ability to construct, evaluate, and iterate machine learning workflows using SciPy, XGBoost, deep learning libraries, and scalable tools (e.g., Cloudera Oryx), including model evaluation and basic deployment considerations.
• Evaluate and document best coding practices—including indentation, commenting, low-level optimizations, use of help/documentation resources, and ethical considerations—when producing collaborative, secure, and domain-aware data science solutions (e.g., healthcare support).