Quantitative Data for "ProVoice: Designing Proactive Functionality for In-Vehicle Conversational Assistants using Multi-Objective Bayesian Optimization to Enhance Driver Experience"
This dataset includes:
- All participant questionnaire responses to mental demand, predictability and usefulness.
- All design parameter values (including pareto-optimal values), split into both experiment conditions (Condition 1: Trained LoA [Adjusted according to design parameter value], Condition 2: Fixed LoA [Adjusted according to level of proactive disposition. Design parameter values for LoA were not used during iterations. Normalised score was added during analysis]).
Analysis also uses rCode: Enhanced R Functions for Statistical Analysis and Reporting.
I acknowledge the use of ChatGPT (GPT-4o, OpenAI, https://chatgpt.com/) for assisting with structuring R code for data analysis.