 Thank you. Thank you, Igor. I'm Eric. I'm Chief Information Officer at CRE, the Center for IntelliSyplinary Research. And Irina is also presenting with me. Yes, so hello. I'm Irina Nikolaeva and I'm products manager at CRE and I'm helping with everything that's online tools for learning. Okay, so what is CRE? Briefly, we are both a department of the University of Paris with courses from all levels, license, master, doctorate, but also courses for youngsters and even a baby lab. So, working with babies and educators from a younger age and a life from learning. And we're also an association. So we are at the same time, a part of a public university, but also a private association. And our aim is to experiment and promote new ways of learning. And everything we do is linked to the United Nations sustainable development goals. I'm pretty sure you all have heard about them. So it's a very useful framework to organize our projects on learning and it resonates very well with the younger generations of students. Who really want to add meaning to what they learn. If you move to the next slide. Yes. So our pedagogy is really based on projects on learning by doing and the students, the workers teams and their judge usually collectively on the projects that they do. Meaning that they have to work interdisciplinary as well a lot. And I know it's a buzzword, but for us it's it's more a real way of working. And we are trying to empower students to make them actors of the learnings of creating new new part of the curriculum and also to work together as peers but also to be mentors to each other when when they're good at something and for the PhD would help masters who would have the license students. And last thing which is most important in the case of open education resources is that we encourage everyone at CRE to share their learning experience both what they use to learn what they produce to learn are very important. So what we are trying to do is to prototype tools to help our learners learn better together as a community. So one of the most important things we're trying to produce is a system to match learners with learning resources. Learning resources can be obviously courses online or not but people are also very excellent resources to learn you learn more from people than from books usually. So be them be their learner co learners mentors experts. And we want people to share their projects on digital platforms. So they can learn together and they can benefit from the previous projects made by researchers or other learners. And what the systems we want to build will be able to match learners with resources with other learners but also with possibly jobs internships and we're trying to build a common architecture for all for all that. So moving to the next slide. Here you can see some screen shots of some of the applications we we provide for this ecosystem so a system we have designed to let any learner document their own projects very very easily a system on the right side to map knowledge on the proposed recommend resources to the to the learner that's the one we were going to focus on today. And the third one on the bottom of the screen is more about letting people declare the skills they want to learn what they're able to do and to match resources. With the with the learners are to help you find experts that could help you to learn. So I'll let Irina describe the tool. So now we're going to focus on one of the tools that Eric talked about, which is the tool to share resources online and to organize them also. So the UCS small screenshot here that consists of a map that I'll explain in a video what this map is about, and then the resources that were added. So right now I'll start the video. I'm going to present to you we learn. I have downloaded the we learn plugin that is available for Chrome and Firefox. And now whenever I browse the web and I find an interesting learning resource, I can add it to my library and we learn. Here I find the UN page on sustainable development goals. Interesting. So I would like to save it into we learn when I click on we learn. We learn tries to annotate the content of the page with some specific concepts. So these concepts are actually Wikipedia articles. What we learn does it is that it compares the content of the page with Wikipedia articles, and it will suggest concepts that have similar content as this page. So here it found that this page speaks about sustainability and more specifically about sustainable development goals. I can now modify this list of concepts. For instance, sustainable development goals is very specific. I don't need the broader sustainability group, but I would like to add that it has been also done by UNESCO. So I would like to tag it with UNESCO. Here is the UNESCO page. And now I am happy with the ways that this resource is tagged by concepts and I can add it to we learn. If I go into my library, I will see this resource added to my list of resources. A very cool feature of we learn is that we can visualize resources not just like a list, but also on a 2D map. If I click on the Discover button here, I will see this map of concepts that have been used to annotate resources within the we learn community. So these concepts are organized here by similarity. So all concepts that will that will tackle technology will be somewhere in the region around the technology label. And if I go up here, I'll find medicine and science, later society and geography history and so on. So if I zoom in a specific area, I will find more and more of different concepts related to this area. So, for instance, here I'm zooming into the society. And here, let's say that I would like to see what are the resources related to the concept Ikigai. I click on the concept Ikigai and I see here all the resources related to this concept that have been entered by the we learn community. And what's interesting is that we can also not share all the resources but make specific thematic groups. So for instance, in the beginning of the COVID pandemic, we made a COVID-19 pandemic group, and we all together shared our resources that we find found that were quality resources. What I didn't say is that the elevation that you see here is response is proportional to the number of resources in a specific area. So as you see here, most of the resources will be somewhere around medicine. Pretty logically. If I zoom into this area, I will see the different concepts that will appear. For instance, if I'm interested in disease surveillance and I click on it, I will see the resources within the COVID-19 groups that are related to disease surveillance. I can also search in the search box for more free keywords for titles and so on. If I search for COVID, and I will probably find quite some resources here around where a COVID isn't the title or somewhere else. And very, very recently, we have introduced the hashtags. So we will be able to have personal hashtags for different resources and to have personal notes for the resources. This will be public very soon in the coming months. So right now you can go to the WeLearn website www.welearn.cree-parry.org. You can download the plugin and you can try it out for yourself. Thank you. That's it for WeLearn. So this was the presentation of the WeLearn plugin. As you can see, I'm switching back to the presentation mode. Yeah, so right now WeLearn can qualify any online learning resource. So it can obviously qualify the open online learning resources, but it's not limited to it. So we can add a filter that would limit search results to OERs, if you want to. But we could also think about actually being inspired by resources that are not OERs and maybe see if one resources of a specific interest make an open resource out of it inspired by it. Since we are using Wikipedia ontology to view the resources and to annotate the resources with it, it's really broad and covers quite some themes. But however, it also inherits of the biases. Wikipedia is still an encyclopedia that was created by western white men mainly. So there are still some biases in it. Hopefully they will evolve also with time and our tool will evolve when Wikipedia evolves. And for now, the learning progress is also self-declared and deduced from online behaviors. We do not yet have anything to actually track the progress. But WeLearn can be used to find the best open OERs and to share them within a specific group of interest. You can also promote the most useful resources and we also are able to recognize the contributors that share these resources or create these resources in such a tool. We plan also to have a better view of who added what resource and then it will be even easier to recognize the contributors. And with such a tool we can identify the really good online learning resources that are worth being transferred as OERs. I already said that. Just to complete that, the important point is that, for example, national institutions could be very interested in to seeing who are the best teachers who create the best resources. In France, when we are discussing the partnership, we have a partnership with the Ministry of Education and for them to be able to see what are the best resources and how to help the teachers creating them and recognize these resources and make them OER. It would be very, very useful because that's also a new way of promoting the best resources. So we go on a bit just to dive a bit on how it works. As Irina said, we are using Wikipedia as an ontology. And the way it works is that we don't know the whole of Wikipedia, the millions of articles of Wikipedia and all its text, and we crunch this data with a machine learning system. Once we've done that, we have a higher dimension matrix that can be used to analyze any text that we feed it in and to see which Wikipedia articles are closest to the text we feed to the system. And after that, we are able to create a two dimensional map of the whole of knowledge, the human knowledge. If you accept that, Wikipedia is an acceptable proxy for representing every human knowledge bit. And the system is very powerful because it means we are able to see similarities between any type of learning resources be them text videos because videos also have text attached to it. People because we can deduce from their behavior what they're good at or from what they've recorded as personal skills and projects. We can also analyze the text of projects and the skills which are attached by the project participants. So if we move on a bit to explain this, we are also developing a tool which is called skills right now just the code name of the application where people can use the same system to declare what they're good at. So here you can see a profile of someone who says he's good at his competent as a databases and web development and just curious about the space science. And if we move to the next slide, how it works is that the person can select keywords and we will get in Wikipedia all the articles related to this system to the to this keywords. And then the person can hear declare what are the skills he has mastered or wants to master. Next slide. So same thing, but with learning objectives, you could say you're good at educational technology, but you want to become an expert. So you can give yourself goals for for learning. And this can be very useful, very useful addition to we learn because now we can match your profile with resources of things you want to learn. Okay, and we can also match people together by what they have declared as things they want to learn, or things are good at, as well as information derived from their learning behavior if they declared a lot of resources about public policy, we can we will be able to to infer that they are already good at public policy and could be a mentor to someone else. Okay, so if we have still some time we can, we can answer a few of the other questions, I can see many questions in the in the chat. Thank you both Eric and Irina for sharing and for sharing also the link it's super helpful I think that all the participants here are encouraged to download the plugin and play with the biller. It's super impressive thank you for sharing. I think Colin, I would like to start with Colin because he asked a question earlier on. He asked, can crowdsourcing help with licensing. That was the question from from Colin. So, sure. As we tried to highlight in the in the presentation. So, crowdsourcing is very important to qualify the learning resources, and especially OERs because the quality of resources can be very heterogeneous. And even more than the quality is the resource adapted to your particular need your report to your class if your teacher or to yourself. So, also our system would be very interesting because you can use crowd intelligence to qualify the best resources, and we are convinced that leveraging both collective intelligence and artificial intelligence gives the best results. Not, not just AI or not just crowdsourcing but both. Okay, thank you very much and I think we have got time for one more question and there is one from Paola Corti from Milano. Is there a risk that semantic similarity cuts of marginal perspectives or is this taken into account in instructing AI to work with resources or is it possible to include this kind of attention in some way also for single projects. Basically, we anytime you use AI you have a risk of introducing biases in the system. And we, we choose, for example, to use Wikipedia as the basis of our system and we, it means we have the same problem that Wikipedia has because it is mainly created by middle age white males from the western countries, for example. We have a lot of biases, but we can hope that it will improve in time and we can build some safeguards in the system. But obviously it's a concern and it's a research area by itself. Okay, thank you very much. Unfortunately, we have another time. So thank you very much to both of you for your for your contributions both arena and Eric. Thank you for the link into the chat window for additional conversations please visit the connect spaces and also Eric and Irina please upload your presentation slides directly there so that people can continue engaging the video from this session will be available to shortly afterwards. I would like to ask our tech support to please stop the recording.