CSC 333: Computational Data Science

An introduction to the practice of data science: collecting, cleaning, analyzing, visualizing, and communicating with data, using the Python tools that working data scientists actually use.

About the course

I designed this course to fit the needs and resources of Southern Virginia University, where it is the one data science course in the computer science offerings. So it does more than cover the fundamentals. I think of it as the cream skimmed from the top of several mainstream courses: fundamentals of data science, linear algebra for data scientists, statistical machine learning, and natural language processing.

The first part of the semester is tooling: Jupyter notebooks, pandas, Git, regular expressions, web scraping, and data visualization. The middle of the course turns to two methods I have used heavily in my own data-driven modeling research, least squares regression (including regularized regression) and principal component analysis (PCA), along with the probability and linear algebra they rest on. We finish with a brief introduction to natural language processing.

There is no textbook. Students “read the docs”: before each topic they work through a reading guide built on the official Python documentation. The rest of the grade comes from notebook-based tasks, two projects with presentations, and a final project.

Prerequisite: MAT 221: Statistics.

Syllabus

• Syllabus (Spring 2025)

Helpful resources

• Python Data Science Handbook by Jake VanderPlas. Free online, and it covers IPython, pandas, Matplotlib, and scikit-learn in the same order we meet them.
• Documentation we read from: pandas, Matplotlib, seaborn, Beautiful Soup, and scikit-learn.
• Pro Git. The free, standard reference for Git. The first three chapters are all you need for this course.
• regex101. Build and test regular expressions interactively, with an explanation of every piece.
• Kaggle Learn. Short, free, hands-on courses, plus a large library of datasets for project ideas.