Shanshan Yang is exploring how AI chatbot technology can support teaching and learning. To do this she has developed an AI chatbot prototype that supports tutors in the development of online reading lists. The prototype chatbot provides answers to technical problems and can give instructions and demos on how to edit and publish digital reading lists. The chatbot helps tutors resolve technical issues, and, in cases where it cannot resolve the problem, directs tutors to where they can get further technical support. In addition to this, the chatbot captures data and uses this to analyse user requirements and create a list of frequently asked questions.
While the prototype chatbot is designed to help tutors, the chatbot technology could be used in the future to help support students, for example by providing instant replies to students who do not live near campus across different time zones. The chatbot could also be developed to offer personalised learning pathways and reduce the workload of tutors and support staff. In the future, the chatbot could be developed to improve accessibility through a built-in voice feature.
Shanshan Yang, Warwick Manufacturing Group
- The first step was for Shanshan to understand the technology. Shanshan used IBM Watson to host the chatbot, and used BotMock to design conversations.
- The second step was to understand the users and how they might interact with the chatbot. For this, Shanshan created dialogue flows using sticker notes of questions that users might ask.
- She then matched these questions with answers and arranged them into logical flows. This was the most challenging part of the process.
- Then Shanshan considered how to host the information. Some of the answers were given in the form of a video, whilst others were given as pdfs.
- Shanshan built the chatbot using IBM Watson.
[In the future], the chatbot could act as a teaching assistant. By answering easy and frequently asked questions it could free up tutors’ time. Also, because some students might not be on campus, this tool could allow them to ask questions anytime and anywhere, in multiple time zones and during holidays and weekends, and still get instant replies.
The level of technical knowledge needed to create a chatbot depends on the use case. The developer needs to be familiar with the task, for example developing reading lists.
If any tutors are interested in developing their own chatbots, I’m happy to share the code!
AI Chatbot Demo
Access the AI Chatbot prototype hereLink opens in a new window.
Shanshan Yang and Kim Stansfield (2022), ‘AI Chatbot for Educational Service Improvement in the Post-Pandemic Era: A Case Study Prototype for Supporting Digital Reading List’, 13th International Conference on E-Education, E-Business, E-Management, and E-Learning (IC4E), 24-29, https://doi.org/10.1145/3514262.3514289Link opens in a new window