First Experience at Nordic-LASI @ 2019

This year CEITER project team together with colleagues from the University of Tartu and University of Oulu organized Nordic-LASI (Learning Analytics Summer Institute) from 28-30 August 2019 in Tallinn University and I got the opportunity to attend this event. In a way, it was important for me as well due to my participation in CrossMMLA workshop organized on the first day of the event.

In this post, I am sharing my experience at Nordic-LASI with a summary of all events. I hope it will be useful for people who could not attend it or who wants to learn about this event.

This year, Nordic-LASI was attended by around fifty participants including senior researchers, data scientist, Ph.D. students, and Master students working in the area of Learning analytics or looking forward to initiating their research journey in this field.

Day-1 (28th August)

Participants started gathering around 1:30 pm and the event started from 2:00 pm. The first event was the CrossMMLA workshop jointly organized by Luis P. Prieto (Senior Researcher @ Tallinn University), María Jesús Rodríguez-Triana (Senior Researcher @ Tallinn University), Daniel Spikol (Associate Professor @ Malmö University), and Inge Molenaar (Assistant Professor @ Radboud University). I and my Colleague Reet Kasepalu (Phd Student @ Tallinn University) also got the opportunity to be a part of the workshop organizing team.

Luis P. Prieto (in below image) presented the idea of the workshop and provided a brief introduction to Multimodal Learning Analytics (MMLA). The presentation slides can be found here.

(Image courtesy: Piret Rani)

After this short presentation, participants were divided into four groups in such a way that each group had at least one technical person for assisting the rest of the group members in data collection as the part of CrossMMLA workshop. Following this, we took participants to the room where we set up the data collection devices for recording group conversation. Once all participants arrived in the room, we provided guidelines for group-activity and for starting the audio data recording (one group during the activity in the below image). It was my first experience of being part of the workshop organizing team. I must admit that I was nervous as well regarding the proper functioning of the setup of data collection. At that moment the only thought I had in my mind was that everything should work in order.

After the first 20 minutes when everything got settled, groups started their activity and recording the data as well. We were using microphone array for audio recording and MQTT protocol for transmitting that data to the server. Additionally, we also used Etherpad tool for allowing participants to share and create the artifact together. Forgot to mention, one of our hardware setups did not work because I forgot to install some libraries. Yes, these things happen that's how we learn. In addition to our data collection prototype, Daniel Spikol also used some IR badges (shown by Daniel in below image) and Kinect to collect some data during the session.

(Image courtesy: Johan Leoste)

Finally, at the end of the workshop session, I took a sigh of relief that we collected data successfully. It was a great opportunity for me for learning and I was really happy to see participants involvement in their corresponding groups for the given activity.

After the workshop, Barbara Wasson (Professor @ University of Bergen) and Tobias Ley (Professor @ Tallinn University) (below image) welcomed the participants and open the Nordic-LASI event. Following it, Jonna Malmberg (Post Doctoral Researcher @ University of Oulu) introduced the keynote speaker to participants.
(Image courtesy: Piret Rani)

The keynote was delivered by Inge Molenaar (in below image) on the topic of Where Learning Analytics and Artificial Intelligence Meet to Enhance Self-Regulated Learning: Envisioning Hybrid Human-Systems.

Though I do not know much about the SRL (Self-Regulated-Learning) this session presented a good opportunity to understand it and particularly its influence on the learning. I really like to mention her research work which used dolphin based illustration to show self-regulation progress to students. It was quite interesting to see their approach to visualizing SRL progress.

After the keynote, we were done for the first day and enjoyed some delicious snacks organized for the event (below image).
(Image courtesy: Piret Rani)

During snacks time, one social event also organized involving people with similar research interest to have an informal talk. I was in the group with Inge Molenaar, Jo Wake (Researcher @ Norway Research Centre), and Tiina (Phd Student @ University Of Oulu). I came to know about Tina's work who is also working in collaborative learning using EDA and Video data to investigate group's affective aspect. It was really a useful conversation with her. Apart from it, I also got a useful book reference from Inge Molenaar on Theory of Distributed Cognition by Edwin Hutchins and also get to know about Jo's work. Overall, it was a great discussion with them.

Around 7:00 pm I returned from the university and started working for next day workshop (Data analysis).

Day-2 (29 August)

The workshop (CrossMMLA data analysis) started at 9:00 am and Luis gave a short presentation about the datasets and ideas about analysis. In the workshop, three datasets were provided to participants. One by Daniel Spikol, another by Inge Molenaar and third the one we collected on the previous day. Participants were divided into three groups, one with each dataset (Group with Inge's Dataset in below image). The participants were asked to explore the data given to them. In addition to the raw dataset, we also provided an initial visualization prototype for data exploration.

The data analysis session followed by a short discussion over MMLA (below image, Phd Student Rogers Kaliisa offering his perspective on MMLA)

Participants discussed the issues of MMLA. Several ideas emerged from the session, all of them were welcomed and drafted on the board for sharing. Snapshots of the board are given below

After the workshop, we headed for a coffee break. I was in desperate need of coffee to keep me attentive. Well, after this break, there was a workshop on Visualising Learning Data by Iiya Musabirov (Senior Lecturer @ HSE University) (in below image). This workshop was focused on visualization of data using R. For me, the concept of Grammar of Graphics was interesting.

I would definitely recommend the documents shared during this session on various visualization using R (document link). So if you are working with R for data visualization, the aforementioned document will certainly provide some useful information. This session also had a group discussion over learning data's visualization and how it can be further refined (below image).

We then headed for lunch. After lunch, a workshop Supporting Self-Regulated Learning with Mobile Learning Analytics by Mohammad Khalil (Researcher @ University of Bergen) was organised. Unfortunately, I could not attend it because we had to make preparations for PhD Mentoring and Networking session. You can find the workshop slides here. Following is the picture which was taken during the workshop.

(Image Courtesy: Mohammad Khalil)

After a short coffee break, there was a Phd mentoring and networking session. I introduced the objective of the session and my colleague Linda Sillat (in the right image) (Phd Student @ Tallinn University) explained the activity participants were expected to do. We expected actually more people but anyhow we started the session. We had this mindset that even if a single PhD student got benefited from this session, the purpose of the session will be served.
There were two groups during the session, one group focused on Pedagogical aspect and another on Technological aspect. As part of this activity, participants discussed their research work, drawn out major research themes and then found potential connection among those themes.
At the end of the session, it was a great joy for us to find out that some of the students really got some good feedback from the discussion :-).





Following are some photos which were taken during this session. In the below photo, the first group having a discussion over finding common research themes in their group and finding potential connection among those themes.

There was a dinner organized for all participants.

Day-3 (30 August)

It began with the workshop on Using Learning Analytics to Design Appropriate, Student-Centered Feedback by Angeliki Eirini Chounta (Senior Researcher @ University of Tartu). I learned some new things during this session i.e. concept of Assistance Dilemna [1]. This session also included a group discussion where we were given some realistic scenario of students performance based on some LA metrics (LMS access). We were asked to design an intervention (e.g. what and how an automated message can be generated and what should be the content). It really made me think and understand the challenges in the field at least to some extends.

It was followed by a series of presentation as part of Symposium: Understanding and supporting learning processes with different multimodal data channels and learning analytics. The first presentation was given by Mohammed Saqr(Post Doctoral Researcher @ University of Eastern Finland) on Social network analysis of learning log data. It was really interesting for me. Moreover, I also had a discussion with him regarding his work.

Next in the row was the Joanna Malmberg's presentation on the use of multimodal data channel to understand group learning. This work used data from various sources (e.g. physiological, logs, questionnaire, etc) to investigate collaboration among groups. Presentation slides can be found here.

This work was closely related to my research work on using MMLA to detect collaboration. In their work, they used EDA to investigate learning behavior in groups. Some useful references were also provided by this presentation (e.g. Questionaire for collaboration). Next presentation by Tiina, was on using EDA and video to investigate the triggers for the arousal state through EDA (you can access the slides here). The last presentation was given by Inge Molenaar on Self-Regulated-Learning.


The session closing addressed by Tobias Ley where he announced the next Nordic-LASI location (KTH, Stockholm).

It was a great opportunity for me particularly for networking, learning new thing, and gaining the first experience of workshop organization.

References
  1. Koedinger, Kenneth & Pavlik Jr, Phil & Mclaren, Bruce & Aleven, Vincent. (2008). Is it better to give than to receive? The Assistance Dilemma as a fundamental unsolved problem in the cognitive science of learning and instruction. 2155-2160.

Pankaj Chejara

Research Scholar, YouTuber, Blogger, Traveller

Tallinn, Estonia

Subscribe to LearnAITech

Get the latest posts delivered right to your inbox.

or subscribe via RSS with Feedly!