Meet our faculty: Konstantinos (Kostas) Pelechrinis
Since joining the School of Information Sciences faculty in 2010, Konstantinos (Kostas) Pelechrinis has expanded the School’s research and education programs from wireless network security to the broader fields of network science and computational urban science. After successfully completing his PhD at the University of California Riverside, Kostas began teaching SIS courses on Network Management, Computer Networks, and Applications of Networks. His first grant supported his work on maintaining trust in Wireless Networks.
Flash forward, and you’ll find Kostas working on “Models and Metrics of Composite Socio-Spatial Networks” with funding from the Army Research Office (he won the Young Investigator Award for 2105). By now, his work has broadened to analyzing civic and social datasets to create models that can facilitate government decision-making and policy creation in areas such as bike sharing, parking, street fairs, and urban navigation.
He recently made headlines with his work on a subject near and dear to many hearts…American football. Kostas and his team used a data-driven approach to identify key in-game factors—turnover differential and penalty yardage, among others—that directly correlate with winning probability. This analysis found that committing one fewer turnover than the opposition presented a 20 percent gain in winning probability. A 10-yard advantage in penalty yardage correlated to a 5 percent difference.
Kostas and his team then used a probability model to create a Football Prediction Matchup (FPM) engine to compare teams, running 10,000 simulations of prior Super Bowls to determine the 2017 winner. Although the final score did not match Kostas’ projections (his model indicated that the Atlanta Falcons would win), it held true until the end of the fourth quarter! Kostas’ work was discussed and debated in publications and web sites form Pittsburgh to Missouri, and in the International Business Times and Science Daily.
Now that the excitement of the Big Game is over, Kostas and his research teams will turn back to their work on urban informatics and location-based social networks through the Network Data Science Lab. He recently completed a sabbatical leave at CMU’s Computer Science Department (while enjoying time with his new daughter and wife) and he is currently serving as the advisor (or co-advisor) for three doctoral students. Kostas’ work on urban informatics will play a key role in both the educational and research agendas of the School of Computing and Information and we’re all looking forward to his many contributions to the reputation and influence of the new school.
Q & A WITH KOSTAS
- Why did you choose the Computer Science field? During high school, I loved math and physics, so I wanted something that combines them and therefore, I went on to study Electrical and Computer Engineering. My undergraduate program actually sparked an interest in computer science through algorithms, problem solving, and the strict structure required for a computing system to work. So, I decided to continue my studies in Computer Science.
- Even though your PhD is in CS, you work for an Information Sciences school primarily in the Telecom program. What about the iSchool was appealing to you? I like getting out of my comfort zone. I like challenging myself and trying new things. With an undergraduate degree in Electrical and Computer Engineering, which is close to Telecom, coming to teach and research in the Telecommunications program was not something completely new for me. The boundaries between areas, fields and departments are very thin nowadays anyway. That’s why I think that the new School of Computing and Information will be so successful – we’re breaking down organizational and physical barriers to research and learning in the fields of CS and IS.
- Could you recommend a book that everyone in the information field and professions should read? I am not going to suggest a book directly relevant to the information science field, but rather a book from Nobel prize-winning psychologist, Daniel Kahneman: "Thinking, Fast and Slow". Everyone should know the "tricks" our brain plays when we try to make decisions.
- What is one thing you’re really looking forward to as we become SCI? I could make jokes about a bigger office, or better parking, but the real opportunity with SCI is the chance to collaborate with faculty from so many different backgrounds – engineering, psychology, ethics, machine learning, and more. The new school has so much potential to foster collaboration among the SCI faculty and students, allowing them to solve important societal problems.
- We know that you just had a baby…what’s the biggest challenge to being a parent of an infant? The one challenge everyone has in this situation – getting enough sleep! The other problem is that there is not good data for understanding what they need when they cry; maybe computational parenting should be the next big thing in Information Sciences...