Skip to main content

After exploratory experimentation sessions with teachers in Slovenia and Luxembourg, the first days of December were reserved for a session with Irish teachers.

The DALI4US data literacy workshop in Dublin, Ireland was held by Blaž Zupan and Janez Demšar from the University of Ljubljana. It was organised by Oide and attended by 12 teachers.

The topic of the workshop was explaining the basic ideas behind recommendation systems, such as those used to suggest videos or articles in online shops, to present ads, and similar. Can students (or teachers) simulate a system like this in the class?

The activity was based on the Recommendation Systems scenario, but instead of recommending cartoons, the teachers were given cards with posters of 60 sitcoms. Each teacher selected eight of his or her favourites. Next, they had to count how many sitcoms they had in common with the other teachers in the group and mark the three colleagues with the most similar tastes. Then, the group drew a network with nodes labelled by their names, and arrows connecting them to the three persons they identified in the previous step. Finally, they wrote down the sitcoms that the three colleagues liked but that they th emselvesdid not choose. Voila, those were their recommendations!

The teachers also entered their selection into a website, so that the data could be loaded into Orange, a hands-on data analytics tool adapted in the DALI4US project for school use. Then, they replicated the steps they had done manually: computing similarities, finding the three colleagues with the most similar tastes, drawing a network, and giving recommendations.

The activity concluded with a discussion in which the teachers shared their perspectives on how effectively it explained the basic principles behind recommendation systems. They also suggested other scenarios, such as recommending books, and thought about how they could replicate it themselves in a primary school classroom. And the main takeaways? The teacher’s enthusiasm was simply infectious – they were fully engrossed in the topic, which once again showed the importance of making data exploration exciting by, among other things, choosing the right theme for the activity.