Preference modeling has been studied extensively in the literature, and hasapplications in recommender systems and automated decision-making. Theeventual objective of working with preference models is to be able to reasonabout preferences over objects, often referred to as outcomes. In most of theliterature, each outcome is described as an assignment of values to a set ofattributes. Representing and reasoning about preferences over outcomes callsfor efficient preference models. In this thesis, we focus on one such model, Con-ditional Preference Networks (CP-nets). A CP-net is a graphical model thatcaptures the preferences of an individual using a directed graph, with verticesrepresenting attributes and edges representing dependency relations betweenattributes. Information about the preferential dependence/independence be-tween attributes can be leveraged to efficiently order outcomes without exhaus-tively comparing all attributes in a pair of outcomes. In most existing studies,it is assumed that each individual user has their unique CP-net representingtheir preferences. In this thesis, we propose an approach to aggregate thepreferences of multiple users via a single CP-net, while minimizing disagree-ment with individual users. We assume that each user has their preferencesrepresented via a separable CP-net, i.e., a CP-net without any edges betweenattributes. Our goal is to represent the preferences of a group of users using asingle CP-net, referred to as a summary CP-net. We present two algorithmsthat assume all the input CP-nets are separable, with results on correctnessand complexity for each algorithm. We also present a discussion on someimportant properties of CP-nets and the impact these have on our algorithms.
اصطلاحهای موضوعی کنترل نشده
اصطلاح موضوعی
Computer science
اصطلاح موضوعی
Information technology
نام شخص به منزله سر شناسه - (مسئولیت معنوی درجه اول )