Constructive Preference Elicitation for Multiple Users with Setwise Max-margin
نویسندگان
چکیده
منابع مشابه
Constructive Preference Elicitation by Setwise Max-Margin Learning
In this paper we propose an approach to preference elicitation that is suitable to large configuration spaces beyond the reach of existing state-of-theart approaches. Our setwise max-margin method can be viewed as a generalization of max-margin learning to sets, and can produce a set of “diverse” items that can be used to ask informative queries to the user. Moreover, the approach can encourage...
متن کاملA MILP Formulation for Setwise Max-Margin Learning
In max-margin learning, the system aims at establishing a solution as robust as possible. In this paper we extend the idea of max-margin learning to cases where the goal is to produce a set of solutions, instead of a single one. In particular, we focus on the problem of preference elicitation, but we believe our idea could be adapted to other situations. We present a MILP formulation for our “s...
متن کاملOptimal Set Recommendations Based on Regret
Current conversational recommender systems do not offer guarantees on the quality of their recommendations, either because they do not maintain a model of a user’s utility function, or do so in an ad hoc fashion. In this paper, we propose an approach to recommender systems that incorporates explicit utility models into the recommendation process in a decision-theoretically sound fashion. The sy...
متن کاملCoactive Critiquing: Elicitation of Preferences and Features
When faced with complex choices, users refine their own preference criteria as they explore the catalogue of options. In this paper we propose an approach to preference elicitation suited for this scenario. We extend Coactive Learning, which iteratively collects manipulative feedback, to optionally query example critiques. User critiques are integrated into the learning model by dynamically ext...
متن کاملTowards Compositional Design and Evaluation of Preference Elicitation Interfaces
Creating user preference models has become an important endeavor for HCI. Forming a preference profile is a constructive process in the user’s mind depending on use context as well as a user’s thinking and information processing style. We believe a one-style-fits-all approach to the design of these interfaces is not sufficient in supporting users in constructing an accurate profile. We present ...
متن کامل