Data Science Across Disciplines

Session-03: ABSTRACTIONS & MODELS

This week discusses ways of abstracting data. We start by visiting statistics as a means of representing data and its inherent characteristics. The session moves on to discuss the notion of a “model” and visit the different schools of thought within model-ing, as well as a tour of fundamental statistical models that help abstract data and its inherent relations.

The practical part explores processing data and data transformations, summarizing data through descriptive statistics, the case of outliers and a brief overview of robust statistics, as well as investigating relations within different aspects of the data and explore concepts such as correlation, regression, and their relevance within different disciplinary frameworks.