Poyraz: Context‐aware Service Selection Under Deception

Abstract

The increasing number of service providers on the Web makes it challenging to select a provider for a specific service demand. Each service consumer has different expectations for a given service in different contexts, so the selection process should be consumer-oriented and context-dependent. Current approaches for service selection typically have consumers receive ratings of providers from other consumers, where the ratings reflect the consumers' overall subjective opinions. This may be misleading if consumers have different contexts and satisfaction criteria. In this paper, we propose that consumers objectively record their experiences, using an ontology to capture subtle details. This can then be interpreted by consumers according to their own criteria and contexts. We then integrate a method for addressing consumers who lie about their experiences, filtering them out during service selection. We demonstrate the value of our approach through experiments comparing our model with three recent rating-based service selection models. Our experiments show that using the proposed approach, service consumers can select the service providers for their needs more accurately even if the consumers have different criteria, they change the contexts of their service demands over time, or a significant portion of them are liars.

Stereotypical trust and bias in dynamic multiagent systems

“…While several authors have attempted to address the general issue of deceptive reputation providers [Sensoy et al 2009;Yu and Singh 2003;Teacy et al 2006], these approaches involve either learning about the trustworthiness of reputation providers (in a similar manner to learning about trustees), or treating statistically "outlying" opinions as untrustworthy. However, in dynamic societies, the turnover among reputation providers may be high, and so it may not be feasible to build models of individual provider trustworthiness.…”

“…When gathering reputational evidence under bias, trustors should avoid integrating opinions that are deemed to be biased in some way. Unlike approaches which address the problem of deception [Sensoy et al 2009;Jøsang and Ismail 2002], we do not assume that opinions that are divergent from the majority are necessarily inaccurate or deceptive. For example, if agents of a minority type are likely to share similar subjective evaluation functions (i.e., they share a perceptual bias), then those agents may choose to seek the opinions of feature-similar opinion providers who are more likely to provide opinions appropriate for them, even though these providers would be considered outliers by naïve reputation aggregation.…”

“…While addressing such issues remains an open problem, a number of approaches exist. These include approaches based on transitive trust networks [Jøsang [Sensoy et al 2009;Teacy et al 2006], and logical representations of graded trust [Lorini and Demolombe 2008]. However, in order to clearly present our contributions, and without loss of generality, we assume a simple model of reputation here.…”

Learning strategies for task delegation in norm-governed environments

“…It is worth noting that we assume agents communicate truthfully. While the issue of deception remains an open problem, some techniques for addressing this assumption have been investigated [32,36].…”

A Probabilistic Approach for Maintaining Trust Based on Evidence

“…Poyraz (Sensoy, Zhang, Yolum, & Cohen, 2009) is a trust-based service selection approach. Poyraz calculates the estimated trustworthiness of web services based on both direct experience and referrals.…”