David Bañeres

Membre of SOM Research Lab at the Universitat Oberta de Catalunya (UOC)

Introducing the smart tutor

What advantages will a smart tutor give students?
One of the challenges we face at the UOC as a virtual university is providing students with the most personalised support possible. In universities in which the students are present, they can ask the teacher directly, while here it’s done via email or in a virtual classroom. A smart tutor would allow us to move towards what would be an ideal situation: for the UOC student to receive the most personal responses possible instead of in a group, and for the support to fit the problem of each individual student.
Could AI end up replacing today’s real human tutors at the UOC?
A lot of people, including the teachers, worry about machines replacing humans. I think that in education replacing a teacher with a machine would be a mistake. What artificial intelligence has to do is help teachers automate certain repetitive tasks and, at the same time, provide students with support. It should not substitute but rather help both parties.
So, what would this smart tutor consist of?
We want to develop a computer application that helps the students and the teachers with certain task in the university. There are three levels. The first begins when the student is getting information about the options at the university. Depending on your profile and the data of other students compiled by the system, the smart tutor can recommend which degrees, masters or courses best fit the student’s preferences and capabilities. That way something that happens often can be avoided: signing up for a course you think you’re going to like, but later you realise you’ve made a mistake.
Let’s go to the second level, once you’re on campus.
On the campus, data management will make it possible to propose how many subjects the student should take and what they need to do to get through them. And, on the third level, in every course and every subject the smart tutor will help both the student and the teacher. For example, in the forums for the courses every year the same questions come up, and the teacher has to invest the same time each time reproducing the same answers. The smart tutor can analyse what the students want to know and the questions that have come up in previous years and then respond automatically. That saves the teachers time, which they can invest in more productive tasks, such as answering new questions and coming up with resources that could be useful for the subject they teach.
What else can it do?
Establishing a system of notifications that warn both the student and the teacher about whether the student is on course to pass the subject. With AI we can predict, from the beginning of the course, whether the student will pass the subject or not.
What would be the result of these predictions and notifications?
The smart tutor will help to reduce the number of fails and drop-outs. It will do so through making the necessary predictions and warnings as well as personalised recommendations for each student, something the teacher of each subject will take part in, because they are the ones who know what needs to be done to pass.
How will students get the information?
We’ve created a new app where each student can see graphs and percentages of the predictions for each subject and which marks they need in which activities in order to pass. Email will also be used, and chats like WhatsApp, and voice, like Amazon’s Alexa. Firms like Amazon sell their system so it can be integrated into others. That way we won’t have to invest time in creating our own emotional chat bot.
Voice assistants like Alexa have emotions. If you insult them, they respond... politely. A chat bot must seem human and, in our case, the way a student talks allows us to understand their mood and respond in one way or another.
When will the tutor be ready?
We’ve been working on it for six months and we’ll finish it in the 2022/2023 school year. We’ve begun doing tests with thee prediction system. The project is part of the Soul University programme, promoted by the UOC’s eLearn Center, which aims to introduce AI into our university in the next 10 years.

Sign in. Sign in if you are already a verified reader. I want to become verified reader. To leave comments on the website you must be a verified reader.
Note: To leave comments on the website you must be a verified reader and accept the conditions of use.