For a better experience, click the Compatibility Mode icon above to turn off Compatibility Mode, which is only for viewing older websites.

Drexel CCI Faculty Spotlight: Julia Stoyanovich, PhD, Assistant Professor of Computer Science

Julia Stoyanovich

December 4, 2017

Julia Stoyanovich, PhD is an assistant professor of computer science at Drexel University’s College of Computing & Informatics (CCI) whose focus lies in the area of data and knowledge management.

Stoyanovich, who joined Drexel’s faculty in 2012, is also a key member of the College’s recently launched Women in Computing Initiative, and recently moderated a CCI Corporate Partners Program panel discussion on November 27 with industry members and alumni, speaking on issues affecting women in STEM fields.

Prior to joining academia, Stoyanovich spent five years in the start-up industry as a software developer, data architect and database administrator. This experience motivated her “to work with real datasets whenever possible, and to deliver results of [her] research to the communities of target users, as part of open-source systems or as stand-alone prototypes.”

Stoyanovich is especially interested in responsible data management and algorithmic accountability, an issue that she and CCI undergraduate and graduate students have pursued in her research group, Drexel Database Group. Together with her colleagues from the University of Washington, University of Michigan and University of Massachusetts Amherst, she recently received $1.6 million from the National Science Foundation’s BIGDATA program, to support research on Foundations of Responsible Data Management. Stoyanovich is the lead Principal Investigator on this project. She is also a member of the ACM Code of Ethics 2018 Task Force, and of the Community Principles on Ethical Data Sharing taskforce, an initiative launched by Bloomberg, BrightHive and Data for Democracy that aims to develop a Data Science Code of Ethics. She is also a steering committee member of Conference on Fairness, Accountability and Transparency (FAT*).

Stoyanovich spoke with CCI about her beginnings in computer science, and shared her thoughts on increasing diversity and inclusion in STEM, and on the importance of algorithmic fairness and transparency in today’s society.

CCI: How did you become interested in the computer science field?
Julia Stoyanovich: I was always interested in mathematics and computer science. I grew up in Russia and my grandfather was a programmer in the Soviet space program, he calculated trajectories of the Sputnik [Sputnik I was the first artificial Earth satellite launched by the former Soviet Union in 1957], so I was exposed to these concepts from a very early age. He never really talked about his work, but he always had this passion for math and for computation, which he imparted on me.

CCI: When did your education in computer science begin?
JS: 
When I was 14, my family moved to Serbia (part of the former Yugoslavia), where I went to Belgrade Mathematical High School, a magnet school that specializes in math, physics and computer science. I was lucky to land in an environment where I had access to different programming paradigms and languages from an early age.

I came to the US to finish 12th grade, when the war broke out in Yugoslavia, and then went to college at UMass Amherst. I wasn’t sure whether I should study math or computer science, and so I ended up majoring in both. But computer science right away interested me more because there’s just so much you can do. It’s such a young discipline. To be creative in math, you must first internalize thousands of years’ worth of concepts that people have developed. With computer science, it’s really only 50 to 60 years’ worth of material.

CCI: What was it like as young woman studying in a predominantly male field?
JS: There were very few women in my college program. As far as I can recall, there wasn’t a “Women in Computer Science” organization or anything of the sort. But I was used to having very few women around me because my high school was predominantly male; in my group of 21 students in high school, only six were girls.

CCI: How do you think we can create a more inclusive environment at CCI?
JS: I think it’s extremely important to just see other women around you. At Drexel, we are doing an excellent job at being visible, with groups like the Drexel Women in Computing Society (WiCS) and also within the Computer Science Department, where we have a significant number of female faculty compared to other institutions. I think it comes down to the visibility of mentors, faculty and other students. We need to speak about these [diversity] issues, and to offer courses that are appealing to a more diverse crowd. Computer science, of course, is a technical field. There’s absolutely no question that women are able to do deeply technical and mathematical work equally as well as men.

CCI: What are you most excited about in your teaching and/or research?
JS: I’m very excited to soon be developing a course on responsible data analysis. It’s very much in the public eye these days that algorithms can and do discriminate against individuals and demographic groups. This is because the algorithms that we use for decision-making today, in the public sphere as well as commercially, are often trained on data. What that means is that how an algorithm behaves depends on what kind of data it uses to shape itself. The data sets we have available to train algorithms are biased because our society is biased, so then algorithms will also start making these racist and sexist kinds of decisions. The history of our society is embedded in the data.

Within my research group [Drexel Database Group], we are looking at algorithmic fairness, accountability and transparency. We are particularly interested in operationalizing fairness and transparency in all the steps of the data lifecycle that come before data analysis – in data sharing, acquisition, integration and querying. As a society, we must understand how to reason about bias, and how to make complex data-driven processes transparent. This is particularly important because algorithms are already being used by government agencies, whose mandate is to distribute resources in a way that is equitable and inline with the public interests. When you have an algorithm that enacts policy and does so in a way that is biased and opaque then, essentially, you are jeopardizing democracy. You have to have oversight over policy.

CCI: How can students get involved in the Drexel Database Group?
JS: I advise PhD students and work with master’s students and undergraduates (through independent study projects, part-time research assistantships and full-time coop). I’d most definitely encourage undergraduate students with the necessary technical skills [third year and up] to get involved.