SEED projects 2020 – Designing for learning

To ask or not to ask: Chatbots in education

Dr Dulani Jayasuriya

Dr Dulani Jayasuriya and Prof Kevin Dow (Faculty of Business and Economics)

Artificial Intelligence (AI) is changing the way students learn and absorb information thereby enabling educators to provide a personalised learning environment. To aid in the learning process, we aim to develop a chatbot built to enhance meaningful student interaction, collaboration and contribute to the innovative education-technology world.

A popular application of artificial intelligence is intelligent tutoring systems that provide a personalised learning environment to the students by analysing their responses and how they go through the learning content. Similarly, chatbots with artificial intelligence technology can be used to teach the students by turning a lecture into a series of messages to make it look like a standardised chat conversation.

Our chatbot work in a similar pattern. It uses lecture slides for each lecture topic uploaded onto the learning management system, Canvas. It was programmed to answer the most common questions asked by students on a per slide basis for each set of lecture slides. If required, it also notes areas where students often go wrong per each slide. It shows students questions and answers students in the past have posed to better understand each topic. Finally, after learning each set of slides (each topic) the chatbot automatically quizzes the students on each slide set. This method results in students learning at their own pace, being more inquisitive, and gaining vital information about each lecture topic in one place. It will increase the productivity of students and engages them more. Moreover, the second part of our project was developed especially for students with learning disabilities as we will incorporate a version with enlarged fonts (and other useful disability features) to facilitate inclusive learning.

In addition, this results in reducing the time teachers spend on monitoring students as the bot will enhance student performance of quizzes. This initiative will be most helpful for large classes with students from diverse educational streams. Finally, using machine learning and artificial intelligence allows bots to draw context from a conversation and reply in a way that is personal, engaging, and conversational. And each task that a bot is able to handle instead of an educator allows us as educators more time to focus on their primary role: educating!