Aggregating Reputation Feedback
نویسندگان
چکیده
A fundamental task in reputation systems is to aggregate multiple feedback ratings into a single value that can be used to compare the reputation of different entities. Feedback is most commonly aggregated using the arithmetic mean. However, the mean is quite susceptible to outliers and biases, and thus may not be the most informative aggregate of the reports. We consider three criteria to assess the quality of an aggregator: the informativeness, the robustness and the strategyproofness, and analyze how different aggregators, in particular the mean, weighted mean, median and mode, perform with respect to these criteria. The results show that the arithmetic mean may not always be the best choice.
منابع مشابه
k-Shares: A Privacy Preserving Reputation Protocol for Decentralized Environments
A reputation protocol computes the reputation of an entity by aggregating the feedback provided by other entities in the system. Reputation makes entities accountable for their behavior. Honest feedback is clearly a pre-requisite for accurate reputation scores. However, it has been observed that entities often hesitate in providing honest feedback, mainly due to the fear of retaliation. We pres...
متن کاملTransactional Security for a Distributed Reputation Management System
Today, reputation systems such as ebay’s prominent ”Feedback Forum” are becoming more widespread. In such a system, reputations are formed by aggregating ratings participants give and receive. These reputations, however, are bound to a specific platform preventing participants from taking and showing their hard-earned reputations elsewhere. That makes the reputations less valuable and leaves th...
متن کاملInformativeness and Incentive Compatibility for Reputation Systems
Reputation systems, which rank agents based on feedback from past interactions, play a crucial role in aggregating and sharing trust information online. Reputation systems are used to find authoritative web sites and ensure socially beneficial behavior on auction sites. The main problem faced by reputation system researchers is a lack of good metrics for comparison and evaluation. This thesis d...
متن کاملThe Dimensions of Reputation in Electronic Markets
We analyze how different dimensions of a seller’s reputation affect pricing power in electronic markets. We do so by using text mining techniques to identify and structure dimensions of importance from feedback posted on reputation systems, by aggregating and scoring these dimensions based on the sentiment they contain, and using them to estimate a series of econometric models associating reput...
متن کاملTree-trust: a Novel and Scalable P2p Reputation Model Based on Human Cognitive Psychology
The key to the success of a P2P reputation system is the feedbacks aggregating mechanism. Based on human cognitive psychology, Tree-Trust, a novel reputation model is introduced, in which the concept of direct trust tree (DTT) is presented innovatively. The main contributions include: (1) in Tree-Trust, feedbacks are searched by using DTT instead of in broadcast way of other work, which makes t...
متن کامل