Towards Enhancing Trust and Accuracy in Preference Elicitation (PDF)

Systems which elicit preference from users are presented with a unique set of problems based on their nature. This paper examines user trust as a possible reason why preference elicitation systems are met with resistance from users. With the example of a personalized news recommendation system in mind, we examine ideas control, understanding and personalization. We make recommendations for improvement to this system and the subsequent study that should see increased improvement in user satisfaction and system accuracy. The underlying goal is that increasing the user’s trust in the system, which will in turn increase the accuracy of preference elicitation.