[Home] [Top] [Archives] [About] [Options]

OLDaily

Welcome to Online Learning Daily, your best source for news and commentary about learning technology, new media, and related topics.
100% human-authored

Presentation
Transforming Learning Design in the AI Era
Stephen Downes, Jul 13, 2023, Transforming Education in the Artificial Intelligence Era, Rabat, Morocco, via Salle de conférence AI et Transformation pédagogique


In this presentation I consider some potential applications of AI in the design of online learning (and MOOCs in particular), point to the importance of data in these design processes, and discuss how data development and quality will become of major importance to learning design in the future.

[Slides]


Considering Large Instance Federation
Woof.group Announcements, 2023/07/13


Icon

This is a smart and well-reasoned article considering the ramifications of "smaller Fediverse instances like woof.group will be able to interact with much larger instances. Threads now claims over a hundred million accounts." Woof has about 1,000 members, which makes interaction a challenge, but Woof also serves LGBTQ+ leather people, which offers opportunities for interaction with larger communities of the same people on (say) Instagram and Tumblr, but also creates concerns about how they will be moderated and how they will moderate others. I can't really summarize the whole discussion here, but as I said, it's quite smart, and well worth your time reading. If it works for you, substitute your favourite alternative community for Woof in the discussion - Mormons, say, or socialists, or environmentalists, or herbalists, or astrologers, or.... whatever. All these considerations apply. And note well: understanding how we manage such differences in the fediverse is key to understanding how we manage them in society at large, because like it or not, we all interact. Via Boris Mann.

Web: [Direct Link] [This Post]


Data-Centric AI: the Systematic Engineering of Data to Build AI Systems
Irving Wladawsky-Berger, 2023/07/13


Icon

In my presentation in Lleida last year I talked about the connection between quality data and AI. This column addresses this topic with reference to Data-centric Artificial Intelligence: A Survey, a June, 2023 paper by researchers from Rice University and Texas A&M, albeit after several paragraphs of boring luddite-bashing. But to the main point: "Data-centric AI systems shift the focus from models toward data. Data is viewed not merely as fuel for AI, but as a determining factor in the overall system quality, and a way to help build AI systems that deal with complex real-world problems." AI systems like chatGPT might not be very reliable at the moment, but the expectation is that with the right data, they will become that much more reliable.

Web: [Direct Link] [This Post]


Data integrity vs. data quality: Is there a difference? - IBM Blog
Sajan Kuttappa, IBM Blog, 2023/07/13


Icon

This article could be more clearly written but it nonetheless highlights some important concepts for any data-driven organization. In it, 'data integrity' is defined as "the overarching completeness, accuracy, consistency, accessibility, and security of an organization's data." In turn, 'data quality' is "essentially the measure of data integrity." So, in essence, 'data integrity' asks, can you measure these important aspects of data, while 'data quality' asks what are those measures.

Web: [Direct Link] [This Post]


WordPress Playground
WordPress, 2023/07/13


Icon

This is a brilliant demonstration of the future of the web (not to mention a really practical tool for WordPress developers). It's an entire WordPress site that runs in your browser. Not just the web pages that you would normall see, but all the editing functions, templates and plugins that adinistrators run in the back end. Here are the details: "Playground uses the latest, cutting-edge technologies to make the key WordPress dependencies work in JavaScript. PHP runs as a WebAssembly binary. MySQL is replaced for SQLite via a WordPress plugin. Web server is implemented in JavaScript as a Service Worker."

Web: [Direct Link] [This Post]


Revolutionizing Online Learning: The Potential of ChatGPT in Massive Open Online Courses
Mohamed Alseddiqi, Anwar AL-Mofleh, Leena Albalooshi, Osama Najam, European Journal of Education and Pedagogy, 2023/07/13


Icon

It was only a matter of time, of course, before a paper like this was published. Despite the claims of sceptics, MOOCs have not gone away, and from the earliest days the potential AI in MOOCs has been apparent, with Norvig and Thrun using AI to support mass assessment in their ground-breaking course on machine learning. But what about 2023's newest arrival, chatGPT? This paper (5 page GPT) reports on "a series of experiments to test the ability of ChatGPT in responding to questions from different subject areas and varying levels of difficulty within two well-known MOOCs platforms, edX and Coursera." Accuracy ranges were good, but nothing like what would be needed to apply it in production courses.

Web: [Direct Link] [This Post]


Learning Analytics for Peer Assessment: A Scoping Review
Kamila Misiejuk, Barbara Wasson, The Power of Peer Learning, 2023/07/13


Icon

This paper (22 page PDF) is one of 27 from the open access volume The Power of Peer Learning. It points to some practical implementations of learning analytics (LA) in peer assessment (PA) including "feedback classification, a text analysis of rubric answers, combining peer and automated assessment, predicting the accuracy of peer raters, text analysis to monitor feedback quality and appropriateness, and clustering and visualisation techniques to optimise the feedback process." As a scoping review, this paper seeks to establish where LA in PA is actually being used, and what we can learn from that. The authors "found that most research focused on addressing the challenge of scaling PA, developing new PA tools enhanced by analytics, or attempting to inform PA theory by evaluating different types of PA."

Web: [Direct Link] [This Post]


We publish six to eight or so short posts every weekday linking to the best, most interesting and most important pieces of content in the field. Read more about what we cover. We also list papers and articles by Stephen Downes and his presentations from around the world.

There are many ways to read OLDaily; pick whatever works best for you:

This newsletter is sent only at the request of subscribers. If you would like to unsubscribe, Click here.

Know a friend who might enjoy this newsletter? Feel free to forward OLDaily to your colleagues. If you received this issue from a friend and would like a free subscription of your own, you can join our mailing list. Click here to subscribe.

Copyright 2023 Stephen Downes Contact: stephen@downes.ca

This work is licensed under a Creative Commons License.