Place recommendation modules in your repository content, create new discovery experiences, and engage your audience with the most relevant content from across all of the sites in our OER Network.
We have designed rich models of users on OER sites, and used these models for recommendation and personalization of learning materials. More specifically the user modelling architecture supports real-time, cross-site and cross-lingual user models and recommendation techniques which take into account the user and content meta-data available in online learning environments.
On top of this we have developed a global cross-site and cross-lingual recommendation engine. It uses machine learning techniques for their core, and semantic technologies to ensure valid combinations of recommended materials and existing skills of the user.
Cross-site recommendation uses content models of learning materials to identify related and complementary learning materials between sites. Cross-lingual aspects are handled by using Wikipedia, and comparing text documents across top 100 Wikipedia languages.