Page 53 - Trust in ICT 2017
P. 53

Trust in ICT                                                1














































                               Figure 5 – Reputation bootstrapping using an adaptive approach

            Trust bootstrapping is a mechanism to assign trust rate for a new service that its trustworthiness is unknown
            and before having any requestor interacting with it. Trust goes through three development phases: trust
            building, stabilising trust, and dissolution [20]. Most studies assume a system where trust already exists (i.e.
            stabilising trust phase). However, it is important to initialise trust rates for new services and service providers
            (i.e. building trust phase). Building trust phase is a crucial stage in any trust relationship. Trust bootstrapping
            is the first step in the trust building development phase and the important step in the trust establishment
            process. It is important to establish trust for service providers and select a service based on its provider’s
            trustworthiness in addition to the service’s own trustworthiness. The trustworthiness of a service provider
            can enhance the requestor’s trust in its services. A requestor can select a service from providers of the highest
            level of trust. Considering trustworthiness of service providers supports trust bootstrapping the providers’
            new services. For example, if a provider is known to be trustworthy, the requestors will trust the provider’s
            services and encourage to select its new services.
            A low initial reputation is assigned if the rate of maliciousness (ratio of defective to total transactions) is high,
            and high initial reputation is assigned otherwise.

            6.2.6   Trust Computation
            The goal of this sub-section is to provide a brief idea about the existing strategies available in the research
            literature and identify the vital points that needs to be addressed and enhanced.
            The paper [21] suggests the combination of trust, mobility and QoS estimations to provide a more reliable
            and rewarding pervasive service experience in MANET. The decentralized trust management model allows
            the dynamic calibration of the service selection, based on a history of service provisions; this should in turn
            promote co-operative behaviours among the various peers. An effective lightweight metric needs to be
            devised to allow communication of expected future movements, a subject of further work.



                                                                                                           45
   48   49   50   51   52   53   54   55   56   57   58