The I-TUTOR project team would like to show you the creative process behind the project work. We conducted short interviews with various contributors to offer you a unique view on the I-TUTOR development. Our first interview is with Dr. Karsten Lundqvist from University of Reading, UK.
What is your role in the I-TUTOR project?
I am a Lecturer in Systems Engineering at the University of Reading (UoR), and I am the principle investigator from UoR within the I-TUTOR project. UoR has got several tasks within the project, however our primary work concentrates around the implementation of different tools. For instance we have made the statistical data mining aids and the chatbot system. We have also designed the website of the project.
What was your main interest in the project?
Personally I find it very interesting to investigating and designing AI tools within educational settings. Work has been done in the area, but not enough, and the possibilities of using these tools are plentiful. In the past many disregarded them because it is difficult to predict the benefits, but I think that a project like I-TUTOR allows us to look at this pragmatically. That is what is needed.
What did you find suprising, more complicated or easier than expected?
When developing data mining tools I always see myself as a “detective”. I am investigating data to try and find common features or indeed outliers within the pool of data. We had some pools of data to begin with, however the structure of the data from the different pools were so significantly different that it was impossible to find commonalities. Educational data might be like that, due to the different styles of teaching/learning that happens in real life. Therefore many standard data mining tools could not be used to develop the I-TUTOR tools.
The pragmatic approach was then to create a tool that tutors and students can use to highlight the current situation that they find themselves in. This is what the statistical datamining tool aims to achieve. It was hard to get to that, because my personal hope was to utilise more AI techniques in this area.
What new questions arise while working on the project?
It would be interesting to investigate bigger educational data samples to see if the diversity is common across the sector, or if there are commonalities that could be identified at a higher level when investigating across the sector. However we are talking many data sources over several years. Such a project would have to be big, and involve many more partners than I-TUTOR has involved.
What do you think about the future of the results, what would you like to see as a continuation of the project?
I think that the pragmatic approach we have taken is a good approach that are producing tools that can be used in real educational settings. I am confident that they will be used. This in itself is a very good result. I would also like to see new research questions being addressed in future project. This is a fantastic research team, and it would be a shame not to continue the work that we have already achieved.