Print this page

Demystifying data: An expert clears up the confusion on today’s hottest jobs

Written by 
  • July 12, 2021

By Yuliang Zheng, Ph.D.

Chair, UAB College of Arts and Sciences Department of Computer Science

rep data science screen 550pxPut “data,” “analytics” or “AI” in the title of any job and suddenly it seems much more desirable. Between 2016 and 2019, data scientist held the top spot in Glassdoor's annual ranking of the best jobs, which is based on a combination of median salary, job satisfaction and job openings. (After slipping to No. 3 in 2020, it climbed to No. 2 on the 2021 list, behind Java Developer.)

The U.S. Bureau of Labor Statistics (BLS) projects job growth for all "mathematical science" careers at 27.9% through 2026. As the BLS notes, a master's degree is the typical entry-level education in this field.

At UAB, our Master of Science in Data Science program was the first of its kind in the state when it launched in 2019 and already has the highest enrollment of any graduate program in the Department of Computer Science. But do you need a master’s to work with data? Do you need to know about machine learning? And what exactly are employers looking for? As the chair of the Department of Computer Science at UAB and a practitioner with 30 years of experience working with banks, startups and government agencies at all levels, I aim to answer these questions and offer a roadmap through a sea of related, consuming terms, with two or three key words you can associate with each.

Data science

Key words: design, solve

Let’s start with data science, which is probably the phrase you will hear most often. Data science is a subfield of computer science. Data scientists design efficient algorithms and other computational techniques for the collection, cleansing, organization, storage and analysis of typically massive datasets (“big data,” which we will come to shortly) that arise from real-world applications. The goal is to use those algorithms to build efficient computer software and hardware systems for

  • Solving real-world problems, or
  • Assisting in decision making.


There are many different titles in data science and related fields, but these are the major ones:

rep data science yuliang zheng vert 300pxYuliang Zheng, Ph.D., chair of the Department of Computer ScienceData scientist jobs requires at least a master’s degree in data science or computer science, and preferably a doctorate in data science or computer science. One unique aspect of the UAB Master of Science in Data Science program is the ability for students to work directly with the university’s world-class researchers in life sciences, natural sciences, social sciences, engineering and other fields on cutting-edge problems related to real-world applications or scientific discoveries. Whatever the topic, a data scientist’s job is to discover or identify the best algorithm for solving a problem. Current average salaries for data scientists are $120,000.

Data engineer jobs typically require a bachelor’s or master’s degree in data science or computer science. A data engineer’s job is to implement, test and maintain algorithms handed over to them by data scientists so that software systems actually solve problems efficiently. Current average salaries for data engineers are just over $90,000.

Data architect jobs typically require at least a bachelor’s or master’s degree in data science or computer science. Data architects design computer systems to collect, organize, transport and store massive datasets. They also work with computer and network engineers to build those systems. Current average salaries for data architects are $121,198 per year.

rep data science graphic full

Data analytics

Key words: apply, understand, predict

Data analytics is the marriage of data science with specific application domains, such as business, health, social media or STEM fields. Data analysts apply data science algorithms and other computational techniques to analyze massive datasets from various application domains, with the aim of making sense of available datasets and, more important, using knowledge gained in the analytical process to help predict future events. Data analytics may also be called data analysis or big data analytics.

rep data science bus analytics 550pxBusiness analytics is one form of data analytics. It is an interdisciplinary field of data science and business focused on applying data science algorithms and other computational techniques in the analysis of business data with the aim of making the best business decisions.


Data analyst jobs require a bachelor’s or master’s degree in data science or computer science, or a bachelor’s/master’s degree in business analytics. Data analysts use software tools for data analysis and interpret the outcomes. They can be found in many fields, from business and STEM to social sciences. The current average salary for a data analyst is $57,261 per year.

Business intelligence analyst jobs require a bachelor’s or master’s degree in business analytics. These positions are similar to data analyst jobs, but they are focused on business only. The current average salary for a business analyst is $66,000 per year.

Big data refers to massive datasets that are too large to be handled by traditional software such as Microsoft Excel spreadsheets or SQL databases. These datasets arise from real-world applications, ranging from financial services and the retail industry to biomedical science, health care, social media, STEM fields and many more.

Sometimes people use “big data” to refer to data science and data analytics, because both of these fields typically work with massive datasets.

Artificial Intelligence

Key words: understand, imitate

Artificial intelligence, like data science, is a subfield of computer science. It is focused on understanding how humans think and act, and on designing algorithms that can mimic human beings in a rational manner.

Machine learning

Key words: patterns, model, predict

rep data science machine learning 550pxMachine learning is a branch of artificial intelligence. Practitioners use machine learning to automatically detect patterns in massive datasets with the aim of establishing a correct model that can be used to accurately predict future data.

There are three forms of machine learning: supervised, unsupervised and reinforcement. In supervised learning, the algorithm gets some external help, such as from a human teacher who provides the correct “answers” or labels for input data. In unsupervised learning, by contracts, the algorithm does not receive any help from an outsider. Reinforcement learning differs from both supervised and unsupervised learning in that the learning algorithm has a feedback loop through interactions with its surrounding environment that rewards or penalizes the learning algorithm based on its performance.

Deep learning is a specialized but powerful subset of machine learning algorithms that use layer upon layer of artificial neural networks — thousands of layers, potentially — which are inspired by the function of neurons in the human brain. The “deep” in deep learning refers to these many layers. Deep learning requires a lot of computing power; it was made possible due to advances in computing technology in the past decade. UAB is home to the advanced Cheaha supercomputer, one of the fastest in the Southeast, giving students, researchers and facultyaccess to the latest hardware for deep learning research.


Machine learning engineer jobs require a master’s degree or Ph.D. in computer science or data science, with specialized training in machine learning. The current average salary for a machine learning engineer is about $111,000.

Putting it all together

Data science and machine learning overlap significantly. Data science relies on machine learning algorithms to make sense of datasets and predict the future. Advances in data science, in turn, help improve the effectiveness of existing machine learning algorithms and discover new ones.