Data Science B.S Frequently Asked Questions (FAQs)

The main goal of the computer science B.S. degree is to provide the theoretical and practical background for someone to become a software engineer. Graduates of the program typically begin their careers working as part of a team that is responsible for some large, existing system, maintaining the code, and making extensions to it.

The data science B.S. program provides students with the knowledge and skills needed to become a data scientist. The typical job entry-level data scientist works as part of a team to build models from data that can be used to understand a phenomenon, and potentially take advantage of this understanding in various ways.

You should choose your program based on which career you are most interested in pursuing—a software engineer or a data scientist.

You might want to consider a double major if:

• You are interested in computer science, but would like to go more in depth into areas such as working with massive data sets, statistical analysis of data, or machine learning
• You are interested in data science, and have a strong interest in the "under the hood" details such as systems aspects and writing/analyzing software
• You are unsure whether you want to start your career as a software engineer or a data scientist

However, please note that you will unless you take a lot more courses, you will have less knowledge/skills than someone majoring in just one or the other. Instead of a double major, another path to consider would be to obtain a B.S. in either computer science or data science, and then get an M.S. in the other.

Most business school data science programs are more focused on providing the knowledge and skills to be able to solve problems in a corporate setting. With this focus, the programs add some components on how to understand a business problem, with correspondingly less focus on the science and technology. This program, offered through the College of Science and Engineering, focuses much more on the scientific, theoretical, and technological aspects of data science.

No, we do not allow data science students to minor in computer science, as the minor coursework is essentially built into the major.

Many of the rudiments of data science have close ties to electrical engineering methods. For example, stochastic gradient descent methods are at the heart of the Least Mean Squares (LMS) method in adaptive filtering (covered in EE 5542). Convolution operations, central to convolutional neural networks, are foundational in linear system theory (covered in EE 2015, 3015, 4541, 5545, 5549), controls (EE 4231, 4233, 5231, 5235, 5721), image processing (EE 5561), and communication systems (EE 4501, 4505, 5501, 5505). Nonlinear optimization (EE5239) also finds applications in signal processing, communications, image processing, power systems, and electromagnetics, to name just a few areas. In this sense, these courses would provide students in the data science program with additional historical perspective and a broader range of application domains for the techniques they would learn in the program.

Data science includes data collection, storage, visualization, and analysis. Although statistics involves all of these areas, it places more emphasis on data analysis and data collection strategies. Statisticians analyze data to learn from it. This typically involves the creation of models to help us understand or predict real-world phenomena. Compared to the statistics major, the data science major places more emphasis on the science of data storage and the associated computer programming.

The required statistics courses emphasize learning from data through the development of models. These models approximate uncertainty or variation in real-world phenomena through probability distributions, which are introduced in STAT 3021 (Introduction to Probability and Statistics). Statistical models for predicting or explaining a numerical output from numerical or categorical inputs are studied and applied in STAT 3301 (Regression and Statistical Computing). Several more statistical models and methods for data analysis and optimal collection are studied and applied in STAT 4051 (Applied Statistics I). The mathematical theory that explains why and how these statistical methods work is introduced in STAT 5101 & 5102 (Theory of Statistics I & II).

The electives focus on specific models and methods for data analysis including interpretation and communication in applications.

Industrial and Systems Engineering (ISyE) is about using engineering thinking to design and operate systems that are efficient, cost-effective, reliable and safe. The focus is on identifying the key levers of the systems that can achieve these goals and on developing a variety of tools to set them in the best possible way.

Because systems can involve a variety of components, including machines and humans, and often operate in market environments, ISyE draws not only engineering but also on business and data science; data being crucial in creating the very models from which recommendations for improvement/optimal operation and design can be derived. 

The focus of data science is focused specifically on learning from data; data manipulation, summarization, visualization, storage, and drawing insights from data. The focus of ISyE is on the design and efficient operation of large-scale systems, e.g. transportation, supply chain, finance, etc. Both data science and ISyE employ methodologies to learn from data and to help people make better decisions.

There are two required ISyE courses in the data science curriculum. In IE 3013, students learn optimization solution methods that drive machine learning. The IE 5533 course will teach students to use data to make decisions using mathematical modeling, in particular, prescriptive modeling.

In general, the ISyE courses approved as data science electives will help students experience, in a relevant engineering context, how data supports an overarching process that starts with identifying system deficiencies and culminates in prescriptions for its improvement.