Practical Data Science

Foundations

CSCI 1109 is a hands-on introduction to practical data science. Students learn how to acquire, clean, analyze, and visualize data in Python, while developing a critical understanding of uncertainty, modeling, ethics, and data-driven decision-making.

The course combines programming, mathematics, statistics, and applied case studies. It begins with Python and pandas, then moves through data cleaning and preprocessing, descriptive statistics and visualization, probability and causal thinking, introductory machine learning, clustering and network analysis, and responsible data science.

Course website

What students learn

By the end of the course, students will be able to:

write and debug Python programs for data analysis using tools such as pandas, NumPy, and Matplotlib
formulate data-driven questions and design analyses to investigate them
apply core ideas from descriptive and inferential statistics
build and evaluate introductory machine learning models for regression, classification, and clustering
create and critique data visualizations for exploration and communication
analyze simple networks using NetworkX
identify limitations, biases, and ethical risks in data science workflows
communicate reproducible analyses clearly

Course structure

The course is organized around a progression from foundations to applications:

Python and tabular data: Python basics, notebooks, pandas, tidy data, grouping, and reshaping
Cleaning and preparation: missingness, duplicates, outliers, feature creation, and preprocessing pipelines
Math, statistics, and visualization: vectors and matrices, descriptive statistics, bootstrapping, plotting, and storytelling
Inference and causality: probability, p-values, A/B testing, confounding, and correlation vs causation
Machine learning: train/validation/test splits, k-NN, regression, classification, clustering, and evaluation
Networks, ethics, and communication: graph analysis, fairness, privacy, responsible data science, and communicating results