Data Science Seminars and Challenge

The course,  titled "Data Science Seminars and Challenge," is centered around the theme of data science and is structured in two parts. Part I: Data Challenge
This component involves a real-world problem, for which data are readily available. Students, working in teams of five to six, will compete to develop a (partial) solution to the problem.  The project spans the Autumn semester and includes the following steps: (1) building a predictive model or methodology based on a training dataset; (2) conducting a blind evaluation of the model or methodology using  test data and a suitable performance metric, which will be used to update a leaderboard throughout the semester; and (3) presenting their approach in a formal oral presentation and submitting a short written report.                                  Part II: Data Science Seminars
This part consists of a series of seminars, all thematically linked to the data challenge project. The opening seminar will introduce the context and present the specific problem for that year’s challenge. Subsequent seminars will showcase various industrial or academic approaches and problems that are loosely related to the challenge’s objectives. Each presentation will last one hour, and students will be expected to read assigned materials in advance to prepare thoughtful questions for discussion after the seminar.