Data Science (BS)

Concepts of problem solving using an object-oriented programming language, algorithm design, structured programming, fundamental algorithms and techniques.

Problem solving by programming with an object-oriented programming language. Introduces data structures. Overview of computer science topics.

Examines quantitative literacy from a data and evidence driven perspective. Looks at the literature behind vaccines, climate, and other contentious topics where there is a wealth of scientific literature and yet these areas are still hotly debated. Investigates ways in which data science is abused; how to mislead with statistics, and how these problems have created a lack of trust in science. Through class discussions, case studies and exercises, students learn the basics of ethical thinking in science, understand the history of ethical dilemmas in scientific work, and study the distinct challenges associated with ethics in modern data science.

Covers the core mathematical topics that underpin data science as well as the key algorithms used for modern data analysis and how to implement them in Python.

Focuses on the exploration of the types of data typically encountered in modern data science, such as text data, spatial data, and time series data. Uses various statistical techniques to gain insight into the structure of the data, including graphical visualization, linear regression, trees and clustering.

Covers the basic statistical concepts underlying data science as well as some of the major methods. Includes fundamental ideas such as the key idea in predictive modeling is the bias-variance tradeoff, and cross validation is the basic approach for dealing with the bias-variance tradeoff. Statistical inference underlies much of data science methodology. Includes Bayesian and frequentist approaches to inference and how they are used in some of key ideas and methods in data science such as causal inference with observational data. Covers some key methods such as K nearest neighbors, naive Bayes classification, A/B testing, linear models, Gaussian processes and data reduction.

Concerns the key modern approaches for uncovering high-dimensional complex structure in data. Topics include regularized regression, ensemble methods with trees, deep neural nets, clustering and dimension reduction and state space modeling.