DSI Student Uses Data Science for Philanthropy
Before enrolling in the master’s program at DSI, Yuriy Loukachev worked for Mastercard’s Center for Inclusive Growth, where he managed the Center’s Data Philanthropy Program. There, he used Mastercard’s aggregated and anonymized data to evaluate trends about economic mobility, social inequality, and income measurements. To unearth meaningful trends and patterns, Yuriy called upon a combination of time-series analysis, machine learning as well as qualitative approaches to analyze the data. His work supported nonprofits, academic research, think tanks and government agencies. In essence, his task was to use data science to promote social good.
Yuriy graduated from the University of Connecticut in 2013 with a double major in Economics and Math. While there, he did research on human rationality and economics. He used game theory to develop models to evaluate how people behave when confronted with economic decisions. Do people behave rationally in a manner that benefits their economic self-interest? Or does irrationality influence a person’s economic decisions and behaviors? While doing this research he was increasingly drawn to data science, which he used to glean patterns about human behavior and to make more accurate assessments about human rationality - and irrationality. The precision of data science helped him understand that what appears like irrational human behavior is often rational.
He enrolled in the master’s program at DSI to hone his knowledge of data science and use it to deepen his understanding of human nature. Here, Yuriy discusses the work he did for Mastercard’s Center for Inclusive Growth; his passion for math and data science; and the work he intends to do after he graduates from DSI.
Can you talk about your work for the Center for Inclusive Growth?
I worked across a spectrum of disciplines (e.g., criminal justice, economic policy, urban planning) to support organizations looking for insights regarding their own data by supplementing them with insights from Mastercard’s data. For example, if a partnering university was interested in conducting a study on the economic impact of construction in a certain New York City neighborhood, I would test several quantitative approaches to analyze Mastercard’s anonymized and aggregated information to find patterns. And I’d often supplement these insights by analyzing publically available data.
Mastercard was okay with sharing its data?
I’d consult with our data-privacy team to ensure that the final output was in compliance with the Center’s policies. Once the analysis was approved, I then worked with our marketing and content teams to develop a narrative that tied in with the Center’s mission of advancing equitable and sustainable economic growth around the world. Another important aspect of my job was attending conferences and events to build relationships within the growing data-philanthropy world to establish Mastercard as a leader in this new field.
What projects did you work on?
My team worked with the Obama administration on criminal justice reform -- especially how crime can affect commerce, i.e. quantifying the negative impact of crime. We also helped certain counties calculate how much funding to request from federal government for certain criminal-justice programs based on our data analysis.
Another interesting project I worked on is Donation Insights, a data-driven site we launched that provides information on philanthropy and donations. Nonprofits use the site to learn about trends in philanthropy -- when and why donors tend to give and to whom. It was a successful product that gives insights to nonprofits that they may not have otherwise had.
You started off working on Mastercard’s business side. Why’d you move to the data-philanthropy team?
During my first two years on the business side, I supported corporate strategy by setting revenue targets and creating macroeconomic forecasts of Mastercard’s performance. It was there that I developed a strong understanding of Mastercard’s data. I moved to corporate social responsibility because I wanted to position myself towards using my company’s resources in innovative ways. I wanted to explore how our data could be used to bridge the information divide between those who are working to solve societal problems and those with the analytical know-how. The data philanthropy program with the Center is an effort to do just that.
How did you first become interested in data science?
My desire to study data science grew from my interest in economics and math. As an undergraduate researcher at the University of Connecticut, I spent a lot of time thinking about rational behavior and its role in economics. I conducted research about modern auction theory and designed a numerical model about how business syndicates form. My objective with the model was to offer advice to antitrust authorities who were evaluating mergers.
Doing economic research experience made me realize that commonplace interactions can be codified and that certain human behaviors could be programmed. I wanted to incorporate data science into game theory and see if people actually acted rationality in their self-interest. I had also developed several models to evaluate how we (if we were perfectly rational!) behave when subject to forces of different markets. With data science, you can gather patterns about human behavior and make correct assumptions about rationality. I realized that so much of economic theory relies on assumptions about rationality, and all this drew me closer to data science, where the use of big data helps make sense of seemingly irrational behavior.
Can you talk about your background?
I grew up in Moscow and moved regularly between the U.S. and Russia when I was a boy, from nine to sixteen. It was really tough as a kid being torn from my friends in both directions, but I reflect on this now and appreciate that this has shaped me in many ways. I think traveling a lot and moving to different states in the U.S. helped me know how to adapt to change. One thing I recognized at a young age was the gap between the American math education system and the one in Russia. Every time I returned to Russia after having spent six months to a year here, I was quite behind in my math classes. I realized how far behind American students are than their age-counterparts in other countries, and how this could be a long-term disadvantage. Quantitative reasoning is a critical skill that not only informs your ability to work on mathematical problems, but problems that occur in everyday life, such as those involving logic, problem solving, critical thinking, and even creativity. I channeled my passion for math when I participated in a program while working for Mastercard called Girls Who Code. Our focus was on helping middle school girls enhance their mathematical thinking skills, especially as it relates to coding.
Why did you enroll in the master’s program at DSI?
I chose the Data Science Institute because I wanted a traditional education that would give me a foundation in the theory behind data science and computer science. In this way, I felt I would gain a more holistic understanding of the emerging field of data science. Obtaining a master’s in data science was a perfect opportunity for me: I could enhance my data-analytical skills and become involved in data-driven projects that have a significant social impact.
I’d like to bring a fresh perspective to how I can help different sectors use their data. I’m especially interested in applying natural language processing to the qualitative data that many nonprofit groups collect. That data has great potential for good, but it often goes unexplored due to the difficulty of analyzing unstructured data. A nonprofit may collect answers from interviews of people, and have difficulty seeing trends from those interviews since they don’t have the ability to analyze this qualitative data. I hope to return to nonprofit work after I graduate and work as a data scientist who can use data for social good.
--By Robert Florida