A user-centric exploration of transparency, explanations, and trust in multi-model and single-model decision support systems
Year: 2023
Type: dissertation
Abstract: Advancements in artificial intelligence (AI) have increased the demand for interpretable decisionmaking processes.This study explores Explainable AI (XAI) by examining the relationship between machine learning model types and explanatory mechanisms.An ensemble of models-XGBoost, Neural Network, Naive Bayes, Decision Tree, and K-Nearest Neighbour-was developed, offering a balance between transparency and accuracy.The models were evaluated using a car assessment task, where participants made decisions and were then shown predictions from single and ensemble models, sometimes accompanied by explanations like LIME and SHAP.Results show that the ensemble model outperformed most constituent models in accuracy and positively influenced user trust, particularly in scenarios of appropriate compliance and incorrect predictions.While explanations had limited effect on trust, the level of agreement within the ensemble significantly influenced user behaviour.Preconceptions such as familiarity and risk appetite also affected compliance.SHAP's waterfall plot emerged as the preferred explanation type.This research contributes methodologically with a novel ensemble model balancing accuracy and interpretability, and empirically by deepening understanding of human-AI interaction.Practical recommendations are provided for presenting explanations to improve user trust in machine learning applications.First and foremost, my utmost appreciation to my principal supervisor, Dr Theodora Koulouri, whose intellectual rigour, patience, and attention to detail have been invaluable to my research.Her persistent support and insightful reviews have immeasurably enriched both my research and my determination, particularly during the most exigent periods of my doctoral journey.In addition, I am deeply thankful to my co-supervisor, Dr Allan Tucker, for his counsel and support throughout the research process.His expertise has significantly shaped the direction and quality of my work.I owe a debt of gratitude to my family, who have provided unwavering support throughout this strenuous journey.My mother's boundless encouragement and emotional support have been my anchor.Furthermore, I am grateful to my father and my brother for their enlightening discussions and insights, which have continually motivated and enriched my understanding.
Author: Elahi, Md Monjur
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7072152117