|Name||252E19 Political Data Science|
|Department||Institut for Statskundskab|
|Teacher||Matt W. Loftis|
|Course type||Tomplads Ordinær Udveksling|
|Course catalogue id||94356|
Description of qualifications:
<p>This course introduces methods broadly termed ’data science’ -- from automated data collection and processing unstructured data to visualisation and methods for exploiting big data. The substantive focus is on applying these skills to study public administration and politics. The public sector produces vast quantities of data. Managing and analysing this information is crucial for academics who study how government functions, for government employees who aim to improve government performance, and for the media and businesses who rely on public data for their work.</p> <p>Students will leave class with broad skills in automated Internet data collection and data management and a focused introductory understanding of several methods for working with big data. Data analysis skills are crucial for understanding the public sector -- however, collecting, storing and using public data present challenges related to budget constraints, privacy concerns, democratic values and equal treatment under the law. Hands-on training is accompanied by readings and frank discussions on the ethical and practical challenges facing analysts in any field working with public data.</p> <p>Practical description:</p> <p>This course is an introduction to data science and statistical programming aimed at the interests of public administration students. Students from other disciplines are welcome, however. The promise and curse of big data are acutely felt in public admin because government at all levels produces vast information. Using public data requires both practical skills and abstract critical thinking skills to execute theory-driven, ethical analyses. This course touches on both.</p> <p>We begin with methods for acquiring and processing Internet data in various forms. The course covers principles and a toolkit for scraping the open web and using application programming interfaces (APIs). Students also acquire hands-on practice with advanced tools for data management and storage. With data management skills in hand, the course turns to conducting theory-driven and interpretable visualisation and data mining. The course ends with introductions to analysing text as data and to using cases for advanced machine learning tools like deep learning.</p> <p>This course is accessible to students who have completed bachelor's level statistical methods and are interested in using the “R” statistical computing software. More advanced students will also benefit. Practically speaking, students learn to collect and exploit data of many types; generate predictions and new measurements; present attractive, informative graphs and apply some machine learning tools.</p> Two crucial background issues will be themes throughout: 1) ethical use of data and models/algorithms and 2) coping with the challenges and opportunities presented by big data.
We have not yet searched for this course in Blackboard.