Õ¬ÄÐÊÓƵ

Imagine you want to learn more about AI. Where would you start? Most people would begin by typing into Google search "learning about AI". But this will bring up nearly 4 million results! Which do you look at and how do you know it is a good way to start? What materials are the most up-to-date? Should you watch a YouTube video, a TED talk, or follow a free online course? The choice can be overwhelming and this can result in giving up: many people don't get very far when learning online. They might try clicking on the first few links in the Google search results and get disappointed that they are not what they need to get started. Or they might find the online material too dry, too high level, too long or too boring. How can we make learning online for everyone more exciting, stimulating and even exceed expectations?

X5GON was an EU-funded project (2017-2020) intended to help students and the general public learn effectively and enjoyably by providing a personalized route through appropriately prioritized open education resources (OERs) such as talks, lectures, texts, slideshows, and online activities. There are millions of these now freely available on the Web and ever more being added each day. The project aimed to develop an extensive architecture, where state of the art machine learning and recommender algorithms are deployed to crawl and classify these resources so that we can then determine how best to help people learn in a way most suited to them. From an HCI perspective, we are interested in how to ensure the learning experience, when using OERs, is informative, interactive, instructive, pedagogical, at the right level, enjoyable and even sociable. To this end,  a range of novel online learning interfaces were designed, intended to support user reflection and engagement, while at the same time enabling users to keep track of their progress, their frustrations, their boredom and stress levels and their sense of accomplishment. These were then tested out with real people in the wild for topics ranging from machine learning to UNESCO historical sites. The ultimate goal was to enable anyone anywhere to have a learning experience that is scaffolded and structured to their needs and at the right level, and afterwards for them to want to keep discovering more - in a customized but also seemingly serendipitous way, akin to how so many viewers now enjoy the experience of seeing what is going to be recommended next when watching shows, programs, films, documentaries, etc. through Netflix and other entertainment companies. But only better because they would be learning new stuff.

Check out the project website: