Acceptance of repeated automated advice

• Main Author: Yeo, Li Ting
• Other Authors: Zinovi Rabinovich
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
• Online Access: https://hdl.handle.net/10356/163347
• Institution: Nanyang Technological University

Modern life is filled with multiple automated advisers, from GPS navigators to trip-booking services to news alerts. Many of them try to adapt to dynamic preferences of their users, some only allow the user to modify their behaviour or advice preferences. Neither is very successful. In this project, we will try to take another look into why that happens. The conjecture is that the acceptance of advice is changing over time. The project will create a controlled advice environment (in a game form) that will allow to test whether and how the acceptance of automated advice changes over time, whether it depends on performance only or the length of interaction itself, which of these parameters is prevalent. The project can concentrate on the creating the environment only (more programming) and expand into a more research oriented variant (studying behaviour models, comparing their performance, etc.)