[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: first pass on terminology
Given a modest lit search, it's apparent that.. 1. there's some incoherence at the (high) abstract level with respect to (wrt) concepts such as "trust", "reputation", etc. 2. besides determining working definitions for said high-level abstract terms/concepts, the lower-level terms and their definitions will depend on whatever ontology and taxonomy(ies) that we leverage and/or develop. So below's just a starter list of terms culled from three sources for comparitive purposes. To a fair degree, I think we need to decide what papers or other sources we're going to leverage, and then derive and settle on terminology largely from those sources, and adding/embellishing only when necessary. That said, here's some terms gleaned from some papers and the present terminology wiki page, which is based on Daniela's input preso. =JeffH ------------------------------------------- http://wiki.oasis-open.org/orms/Terminology ------------------------------------------- Online reputation mechanisms Online reputation mechanisms, also known as reputation systems (Resnick et al., 2000; Dellarocas, 2003a), are using the Internet’s bi-directional communication capabilities in order to artificially engineer large-scale word-of-mouth networks where individuals share opinions and experiences on a wide range of topics, including companies, products, services, and even world events. (Dellarocas (2005) ) Reputation Systems See Online reputation mechanisms. Reputation Reputation is a concept that arises in repeated game settings when there is uncertainty about some property (the “type”) of one or more players in the mind of other players. (Wilson (1985)) Reputation Score A Reputation Score of a Player (Reputee) on the Type (Criteria) by other players (Reputor) is the subjective probability assigned by the Reputor that the Reputee fulfils the Criteria. (Sakimura (2008)) The following definitions were discussed at the kick-off F2F and are (partially) contained in the input document (Bourges Waldegg (2008)) Reputation reputation is a collective evaluation of an entity based on factual and/or subjective data about it, and is used as one of the factors for establishing trust on that entity for a specific purpose. A reputation is a metric (a score, a rank, a state, a multi-dimensional profile, ...) associated to an entity (a person, a business, a digital identity, a website, a system, a device, a category of devices, a computing resource, ...) or to a tuple [entity, attribute(s)] (e.g. [person,skill]) in a particular domain and at a particular moment in time. Reputation domain (or reputation defining space) the encompassing domain where a reputation is defined (to be refined) Reputation calculator (or reputation engine?) A reputation for an entity is computed using a reputation calculator, based on different types of input data about the entity (available within the domain or imported into the domain). The reputation calculator combines and weights one or more input data about the entity, according to a reputation algorithm and contextual information available at the time of computation. Contextual information (to be defined) Reputation algorithm a domain-specific algorithm for computing reputations. A reputation algorithm is designed taking into account the characteristics of the encompassing domain: topology (centralized or distributed reputation computation), entities to which the reputation is associated, entities that produce input data, entities that consume reputations, available types of input data, type of contextual information available, desired properties of the computed reputation (robustness, fairness, etc.), ... Reputation (input) data the data upon which the reputation is computed. Input data can be of different types, for example: Subjective data about the entity: e.g. ratings and feedback from peers, claims and endorsements Static and dynamic characteristics of the entity: e.g. demographics, preferences Behavior data, stemming from measurements and observations within a system: e.g. logs of entity’s past actions, history of interactions, parametric data “Real world” data about the entity: e.g. background checks, credit history Inferred data about an entity: e.g. text analytics Reputation Management System (to be refined) A reputation management system may include mechanisms for: collecting data about entities (generating data inputs or integrating external data); computing reputations; making sure the system is fair (e.g. provide bootstrapping mechanisms for new entities); performing actions based on reputations (e.g. trust computations, automatic decisions); revoking reputations, allowing entities legitimate control over their reputation and allowing entities to challenge their reputations (governance); making sure the system is not abused (security), making sure privacy of entities is respected (i.e. that the association entity - reputation is only disclosed to authorized parties) ------------------------------ "How to talk about reputation using a common ontology: from definition to implementation", Isaac Pinyol, Jordi Sabater-Mir, Guifré Cuní; Artificial Intelligence Research Institute, Barcelona, Spain; 2007 http://www.iiia.csic.es/~jsabater/Publications/2007-TrustWS.pdf ------------------------------ image - refers to evaluations that agents take as certainty, reputation - refers to other's evaluations and therefore is considered a meta-belief, that is, a belief about other's belief. This brings us some important consequences, since accepting a meta-belief does not imply to accept the nested belief. Assuming that target agent A has some given reputation means that it is reputed with more or less goodness, and that such evaluation circulates on the society, but not necessary implies to share the evaluation itself. So, reputation refers to what is said, not what is true. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ social evaluation and voice entity focus Value Strength Evaluation Voice Target Gossiper Recipient Evaluative Belief image direct experience shared voice shared image reputation ------------------------------ A Survey of Trust and Reputation Systems for Online Service Provision Audun Jøsang, Roslan Ismail, Colin Boyd <http://www.oasis-open.org/apps/org/workgroup/orms/download.php/28303/JIB2007-DSS-A%20Survey%20of%20Trust%20and%20Reputation%20Systems%20for%20Online%20Service%20Provision.pdf> ------------------------------ (Reliability Trust) Trust is the subjective probability by which an individual, A, expects that another individual, B, performs a given action on which its welfare depends. (Decision Trust) Trust is the extent to which one party is willing to depend on something or somebody in a given situation with a feeling of relative security, even though negative consequences are possible. (Reputation) Reputation is what is generally said or believed about a person’s or thing’s character or standing. trust system identity trust - describes the belief that an agent identity is as claimed. trust provider access trust - describes trust in principals for the purpose of accessing resources owned by or under the responsibility of the relying party. Context trust - describes the extent to which the relying party believes that the necessary systems and institutions are in place in order to support the transaction and provide a safety net in case something should go wrong. Delegation trust - describes trust in an agent (the delegate) that acts and makes decision on behalf of the relying party. provision trust - describes the relying party's trust in a service or resource provider. Trust purpose - an overarching concept that that can be used to express any operational instantiation of the trust classes mentioned above. Collaborative filtering systems neighbors recommender systems. collaborative sanctioning Centralised Reputation Systems Distributed Reputation Systems Reputation Computation Engines Belief Models Flow Models ------------------------------ A Multi-Faceted Model of Trust that is Personalisable and Specialisable Karl Quinn, PhD Thesis, University of Dublin, Trinity College, 2007 <http://www.karlquinn.com/PhD/publications/2007/A%20Multi-Faceted%20Model%20of%20Trust%20that%20is%20Personalisable%20and%20Specialisable.pdf> ------------------------------ Trust is a term that is described using many synonyms for trust, or "trust concepts." In computer science the trust concepts used to describe trust include “reputation” [Golbeck & Hendler, 2004], “confidence” or “faith” [Shadbolt, 2002], “credibility” or “reliability” [Golbeck et al, 2003], “competency” or “honesty” [Grandison & Sloman, 2000], and “belief” [McKnight & Chervany, 1996]. Such a diverse set of trust concepts, found across a single body of research, reflects the real world where a wide and varied range of individual subjectivity in what trust entails seems to exist across a large and broad population. --- end
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]