Voting (or rank aggregation) is a general method for aggregating the preferences of multiple agents. One important voting rule is the Slater rule. It selects a ranking of the alternatives (or candidates) to minimize the number of pairs of candidates such that the ranking disagrees with the pairwise majority vote on these two candidates. The use of the Slater rule has been hindered by a lack of techniques to compute Slater rankings. In this paper, we show how we can decompose the Slater problem into smaller subproblems if there is a set of similar candidates. We show that this technique suffices to compute a Slater ranking in linear time if the pairwise majority graph is hierarchically structured. For the general case, we also give an efficient algorithm for finding a set of similar candidates. We provide experimental results that show that this technique significantly (sometimes drastically) speeds up search algorithms. Finally, we also use the technique of similar sets to show that computing an optimal Slater ranking is NP-hard, even in the absence of pairwise ties.

Abstract not found

Reputation systems provide mechanisms to produce a metric encapsulating reputation for a given domain for each identity within the system. These systems seek to generate an accurate assessment in the face of various factors including but not limited to unprecedented community size and potentially adversarial environments. We focus on attacks and defense mechanisms in reputation systems. We present an analysis framework that allows for general decomposition of existing reputation systems. We classify attacks against reputation systems by identifying which system components and design choices are the target of attacks. We survey defense mechanisms employed by existing reputation systems. Finally, we analyze several landmark systems in the peer-to-peer domain, characterizing their individual strengths and weaknesses. Our work contributes to understanding 1) which design components of reputation systems are most vulnerable, 2) what are the most appropriate defense mechanisms and 3) how these defense mechanisms can be integrated into existing or future reputation systems to make them resilient to attacks.

Voting (or rank aggregation) is a general method for aggregating the preferences of multiple agents. One voting rule of particular interest is the Kemeny rule, which minimizes the number of cases where the final ranking disagrees with a vote on the order of two alternatives. Unfortunately, Kemeny rankings are NP-hard to compute. Recent work on computing Kemeny rankings has focused on producing good bounds to use in search-based methods. In this paper, we extend on this work by providing various improved bounding techniques.

Reasoning about agent preferences on a set of alternatives, and the aggregation of such preferences into some social ranking is a fundamental issue in reasoning about multi-agent systems. When the set of agents and the set of alternatives coincide, we get the ranking systems setting. A famous type of ranking systems are page ranking systems in the context of search engines. In this paper we present an extensive axiomatic study of ranking systems. In particular, we consider two fundamental axioms: Transitivity, and Ranked Independence of Irrelevant Alternatives. Surprisingly, we find that there is no general social ranking rule that satisfies both requirements. Furthermore, we show that our impossibility result holds under various restrictions on the class of ranking problems considered. However, when transitivity is weakened, an interesting possibility result is obtained. In addition, we show a complete axiomatization of approval voting using ranked IIA. 1

Voting is a very general method of preference aggregation. A voting rule takes as input every voter's vote (typically, a ranking of the alternatives), and produces as output either just the winning alternative or a ranking of the alternatives. One potential view of voting is the following. There exists a "correct" outcome (winner/ranking), and each voter's vote corresponds to a noisy perception of this correct outcome. If we are given the noise model, then for any vector of votes, we can compute the maximum likelihood estimate of the correct outcome. This maximum likelihood estimate constitutes a voting rule. In this paper, we ask the following question: For which common voting rules does there exist a noise model such that the rule is the maximum likelihood estimate for that noise model? We require that the votes are drawn independently given the correct outcome (we show that without this restriction, all voting rules have the property). We study the question both for the case where outcomes are winners and for the case where outcomes are rankings. In either case, only some of the common voting rules have the property.

High-quality, personalized recommendations are a key feature in many online systems. Since these systems often have explicit knowledge of social network structures, the recommendations may incorporate this information. This paper focuses on networks that represent trust and recommendation systems that incorporate these trust relationships. The goal of a trust-based recommendation system is to generate personalized recommendations by aggregating the opinions of other users in the trust network. In analogy to prior work on voting and ranking systems, we use the axiomatic approach from the theory of social choice. We develop a set of five natural axioms that a trustbased recommendation system might be expected to satisfy. Then, we show that no system can simultaneously satisfy

Ranking systems are a fundamental ingredient of basic ecommerce and Internet Technologies. In this paper we consider the issue of incentives in ranking systems, where agents act in order to maximize their position in the ranking, rather than to get a correct outcome. We consider two di#erent notions of incentive compatibility and several basic properties of ranking systems, and show that in general no incentive compatible ranking system satisfying the conditions exist. However, we show that some artificial incentive compatible ranking systems do exist, satisfying only some of the properties.

Reasoning about agent preferences on a set of alternatives, and the aggregation of such preferences into some social ranking is a fundamental issue in reasoning about multi-agent systems. When the set of agents and the set of alternatives coincide, we get the ranking systems setting. A famous type of ranking systems are page ranking systems in the context of search engines. Such ranking systems do not exist in empty space, and therefore agents ’ incentives should be carefully considered. In this paper we define three measures for quantifying the incentive compatibility of ranking systems. We apply these measures to several known ranking systems, such as PageRank, and prove tight bounds on the level of incentive compatibility under two basic properties: strong monotonicity and non-imposition. We also introduce two novel non-imposing ranking systems, one general, and the other for the case of systems with three participants. A full axiomatization is provided for the latter. 1

Personalized ranking systems and trust systems are an essential tool for collaboration in a multi-agent environment. In these systems, trust relations between many agents are aggregated to produce a personalized trust rating of the agents. In this paper we introduce the first extensive axiomatic study of this setting, and explore a wide array of well-known and new personalized ranking systems. We adapt several axioms (basic criteria) from the literature on global ranking systems to the context of personalized ranking systems, and fully classify the set of systems that satisfy all of these axioms. We further show that all these axioms are necessary for this result. 1