Using Transformers to Generate Wellbeing Questions

Using Transformers to Generate Wellbeing Questions

Trasberg, Markus (TU Delft Electrical Engineering, Mathematics and Computer Science)

van der Maden, W.L.A. (mentor)
Gadiraju, Ujwal (mentor)
Allen, G.M. (mentor)
Lomas, J.D. (mentor)
Erkin, Z. (graduation committee)

Delft University of Technology

Computer Science and Engineering

CSE3000 Research Project

2022-06-22

With the surge of mental health issues during COVID-19, more emphasis has turned towards assessing wellbeing. At the same time, recent advances in AI have shown huge potential in a variety of fields. However, few solutions are available at the intersection of those two fields. This research explores how the use of transformer models like GPT-3 could have a positive impact in the domain of wellbeing and proposes a solution to automatize the survey question creation process. After comparing several GPT-3 question creation methods, it was found that through clever prompt engineering and added context, it can be possible to generate syntactically and contextually correct questions about any specific wellbeing context. In addition, the paper discusses potential ways to assess such questions and offers a demo for a question generation web application.

GPT-3
Transformer
Wellbeing

http://resolver.tudelft.nl/uuid:3b7d01de-8d7f-42d8-b147-54cd58da48ae

Student theses

bachelor thesis

© 2022 Markus Trasberg