Personalizing Recommendations in a Large-Scale Education Analytics Pipeline (Collaborative Research: Pardos)

This pilot project aims to begin to organize the world's digital learning resources to make personalized recommendations to learners that are engaging and effective in increasing mathematics learning outcomes. The project accomplishes this goal by developing crowdsourcing techniques to organize learning resources and by analyzing the online learning activities of the student. Teachers are an integral part of this project. The target audience for this pilot is 7th grade mathematics students and teachers.

Full Description

This pilot project aims to begin to organize the world's digital learning resources to make personalized recommendations to learners that are engaging and effective in increasing mathematics learning outcomes. The project accomplishes this goal by developing crowdsourcing techniques to organize learning resources and by analyzing the online learning activities of the student. Teachers are an integral part of this project. The target audience for this pilot is 7th grade mathematics students and teachers.

This project builds upon the Gooru platform that serves a community of over 500,000 in 140 countries and all 50 US states. The platform uses crowdsourcing by its community to curate over 70,000 collections of free web resources consisting of over 16,000,000 education resources. This project builds upon the Gooru resources by using learning analytics on the user interactions within Gooru to discover the resources that most benefit students. Thus, student resources can be tailored to the individual student to maximally engage the student and improve the students learning. Since the Gooru user owns his or her data, explicit opt-in is required for the sharing of data thus protecting the privacy of students who wish not to share their data. Gooru is open source and free so there are no economic barriers (besides internet access) to using the platform.

PROJECT KEYWORDS

Project Materials

Title Type Post date Sort ascending
No content available.