Please log in to watch this conference skillscast.
In this talk, Philipp Burckhardt will demonstrate how the Heinz College at Carnegie Mellon University monitors user behavior in an e-learning environment. Built on Node.js, React and Electron, the ISLE (Interactive Statistics Learning Environment) project, which is built as a part of his PhD thesis, aims at providing students with explorable lessons on statistical concepts.
Philipp Burckhardt will share insights obtained in the process of developing a session logging system that tracks every click and every action of a given student. From weighing the pros and cons of using cookies, local storage or server-side storage to deciding upon whether to store the collected data in a SQL database or document store, he will reflect on the decisions made and the lessons learned.
Gathering data is only the first step, though. Analyzing the data is a much more difficult undertaking. Since collected data does not lend itself to a simple spreadsheet format but consists of time stamps, unstructured text and the like, cleaning and molding the data into an appropriate format is a delicate task. Join Philipp as he elaborates on the development of an online dashboard allowing professors to visualize the collected data, view summary statistics, and apply advanced machine-learning methods to predict future student behaviour and outcomes.
Find out more at www.isledocs.com.
YOU MAY ALSO LIKE:
- Ceci n’est pas un canard – Machine Learning and Generative Adversarial Networks (Online Meetup on 6th August 2020)
- Web Scraping with GoLang (Online Meetup on 13th August 2020)
- Digital Discrimination: Cognitive Bias in Machine Learning (SkillsCast recorded in June 2020)
- Love the Brain You're In (SkillsCast recorded in October 2019)
Monitoring and Analyzing User Behaviour - Intermediate