Authors: Iryna Nikolayeva, Eric Cherel Institution: CRI Country: France
Topic: Technologies for Open Education Sector: Higher Education UNESCO Area of Focus: Building capacity Session Format: Presentation
Abstract
On any given subject, chances are that thousands of online learning resources already exist. But which ones are fit for a specific learner? How can she find the right one, on her exact area of interest, of the right complexity for her unique needs? Our project aims to provide a solution.
We designed a tool that leverages artificial and human intelligence to qualify semi-automatically any online resource. We link it to concepts based on the full list of Wikipedia articles, therefore covering a broad range of human knowledge. Using this ontology has powerful benefits, as Wikipedia covers millions of subjects in dozens of languages and evolve quickly as new concepts emerge, leveraging hundreds of thousands of contributors. Our system also does a rough evaluation of the complexity of the resources, assessed automatically through its lexical complexity.
This results in an annotated library of resources. This library can be shown on an interactive concept map, allowing the browsing to be fun, visual and organized. This map, along with its resources, can then be shared via a link, or with a given group or WeLearn users. After presenting the tool, we will make a live demo. WeLearn is open source, free to use, and already available online.
Keywords
EdTech, artificial intelligence, collective intelligence, crowdsourcing, online learning ressources, wikipedia
I agree this is an interesting project to promote OER and to help students in curating their own collection of resources. I am looking forward to how this project can bring mentors and learners closer.
Hi Ksenia, thank you for your interest. Indeed, it is something we have in mind for the future but we have not yet implemented accessibility features. We are planning for instance to have a full text mode and to be able to evaluate the accessibility of external resources and filter by it. If you have any suggestions, we will be glad to discuss them. We will remove the “Accessible” inscription from the homepage while it is not done yet. Best, WeLearn team.
Very interesting to note the evolution of WeLearn (I returned here to organize OEG Connect content related to AI which now is getting much more attention than in 2020!)
I note from links that WeLearn has evolved so it is part of the cataloguing of the Learning Planet Institute projects? We would like to learn more, is this available software?