Instructor: Douglas Hanes
Saturday, April 13th, 2019, 11:00am - 5:00pm
Sunday, April 14th, 2019, 1:00pm - 5:00pm
With modern digital applications, it is often easy to gather a lot of data on a given individual or system. Hypothetically, this means that we should know more about that individual; on the other hand, how do we sort through all that data to make predictions or reach conclusions? Machine learning techniques -- like tree-based models, support vector machines, principle components, and clustering -- allow one to consider all of the variables at one’s disposal, while still arriving at a relatively efficient predictive algorithm. Deep learning techniques, like neural nets, have the ability to recursively update predictions, in order to hone in on efficient algorithms in unpredictable ways. These algorithms can be used to identify patterns in a data set or to help predict when similar outcomes are likely to be observed in the future.
In this workshop we will present some common machine learning techniques and practice using them to identify understandable relationships in complex data sets. This course is applicable to anyone who wants to harness the growing availability of observational data to gain insights into human behavior and societal outcomes. In this workshop we will discuss the different approaches of data analysis and apply them to some social justice, environmental, and health data. We’ll also discuss the motivations, promise, and ethics of predictive modeling in the social and health sciences.
Prerequisite: Some knowledge of data analysis and regression.
Required Materials: We’ll use the programming languages Python and R as needed. Both are open-source and freely available.
We have a limited number of scholarships for our workshops. If you need an application, please send an email to firstname.lastname@example.org with your complete name and contact information.
If you need academic credits for the workshops, please contact email@example.com.