Illinois Tech College of Science Announces Upcoming Computer Science Distinguished Lecturers

The Department of Computer Science has announced the next two speakers in its Distinguished Lecture Series. The Illinois Tech community is invited to attend.

Lyle Ungar

Lyle Ungar will speak on “Measuring Psychological Traits Using Social Media” on Thursday, April 5 from 12:45–1:45 p.m. in Stuart Building, Room 111. Ungar is a professor of computer and information science at the University of Pennsylvania, where he also holds appointments in multiple departments in the schools of business, medicine, arts and sciences, and engineering and applied sciences.

According to Ungar, social media like Twitter and Facebook provide a rich if imperfect view of who people are and what they care about. He analyzes tens of millions of Facebook posts and tens of billions of tweets for variation in language use by age, gender, personality, and mental and physical well-being. Among other things, he has found correlations between language use and county-level health data that suggest connections between health and happiness, including potential psychological causes of heart disease. Ungar says similar analyses are increasingly being used for applications ranging from job candidate screening to targeted marketing.

Ungar received a B.S. from Stanford University and a Ph.D. from M.I.T. He has published more than 250 articles, supervised two dozen Ph.D. students, and is co-inventor on 10 patents.

Foster Provost

Foster Provost, professor of data science, professor of information systems and Andre Meyer Faculty Fellow at New York University Stern School of Business, will speak on “Causal Targeting: Outcome Predictive Vs. Treatment-Effect Estimation” on Thursday, April 19 from 12:45–1:45 p.m. in Stuart Building, Room 111.

Provost will present the results of a theoretical analysis and supporting simulation analysis comparing treatment effect estimation vs. simple outcome prediction when addressing causal classification. Using outcome prediction may be preferable to treatment effect estimation, even when the best possible models are used for both approaches and there are no estimation challenges (such as confounding). Specifically, outcome prediction is preferable when positive outcomes are (1) very rare, (2) difficult to predict, and when (3) treatment effects are small.

Provost is former director of NYU’s Center for Data Science and former editor-in-chief of the journal Machine Learning, and he was elected as a founding board member of the International Machine Learning Society. His book Data Science for Business is a perennial bestseller. He has a B.S. from Duquesne University and a Ph.D. from the University of Pittsburgh.