Phenomenal Trust Model

The paper deals with the phenomenal trust modelling. Required terms as trust, trust types, trust values and representation are mentioned. Fundamental description of phenomenal trust formation is presented as a form of impersonal trust which is complementary to interpersonal trust. Phenomenon is defined by the set of its possible exclusive values. The model describes the trust of a subject to particular value of a phenomenon. Proposed formula for phenomenal trust formation covers the factors as product reputation, number of recommendations of product, initial trust value of products and trusting disposition of the subject. The behaviour of trust evolution depending on particular factors is studied. I TRODUCTIO Trust is a unique phenomenon and plays an important role in the relationships among subjects in the communities. In the internet age, the trust among the machines, servers, and network nodes gains more and more on importance. Widening of e-service (Liu et al. 2008), e-commerce (Wang and Zhang. 2008), ebanking, etc., arises the question of human machine trust. Further, trust plays an important role in peer-topeer networks (Wu et al. 2008), ad hoc networks, grid computing, semantic web (Wang and Zhang 2008), and multi agent systems, where humans and/or machines have to collaborate. What is it trust and how it can be described? The acceptance of trust is wide and various explanations are offered (Fetzer 1988); from honesty, truthfulness, confident expectation or hope, something managed for the benefit of another, confidence in ability or intention to pay for goods or services in the future, till business credit. The universal trust definition does not exist. Bulk of definitions comes out from Gambetta’s definition (Gambetta 2000). We will understand trust as a given credit, hope, confidence in ability or intention of some subject to perform to benefit of other subject at some future time. Trust is created not only among the subjects (persons, nodes), but the subject can be perceived as a phenomenon, i.e. another type of trust – impersonal trust (Alfares and Hailes 2000; Alfares 2005) complementary to interpersonal trust; trust is formed towards a phenomenon, e. g. to certain product from a set of products of some kind. Interpersonal trust models used for decision on selection of partners can be formed in several ways. The random selection can be the first choice, e.g. (Fort 2003). Further, probability can be applied (Yu et al. 2001), (Winsborough and Li 2002), (Yu and Winslett 2003) and (Rettinger et al. 2007). The game theoretic approach to modeling trust based decisions is proposed in (Baras and Jiang 2004), and (Sankaranarayanan et al. 2007). Next important concept used in decision support is risk (Josang and Lo Presti 2004). Trust models, e.g. (Wu et al. 2008), (Lifen 2008), (Ryutov et al. 2007) usually deploy merely one or two of the factors which determine trust. Present models cover more factors e.g. (Wang and Zhang. 2008). Each of these factors (e.g. reputation, recommendations, and initial trust) can be modelled as an individual component. We propose a model that tries to integrate more of trust affecting factors, i.e. initial trust, reputation, recommendations, and trusting disposition to form a probability based phenomenal trust model. PHE OME AL TRUST REPRESE TATIO Trust has to be measured for applying in society models of trust. However, some simplifications and limiting presumptions must be done. For examining the trust as a behavioural pattern, some ways of representing and possibly visualizing it must be known. Generally, trust can be quantified by a value from the interval 〈a, b〉, where a, b (a < b) are integer or real numbers. Value a represents complete distrust and value b means blind trust. Other verbal trust levels are represented by values from this interval. Without loss of generality, we will use real values from the interval 〈0, 1〉 as is shown in Figure 1. Generally, mapping function which transforms verbal trust values to values from the interval 〈0, 1〉 is neither linear nor symmetrical. Proceedings 23rd European Conference on Modelling and Simulation ©ECMS Javier Otamendi, Andrzej Bargiela, José Luis Montes, Luis Miguel Doncel Pedrera (Editors) ISBN: 978-0-9553018-8-9 / ISBN: 978-0-9553018-9-6 (CD) Trust Value Representation