Trust Evaluation in Service-Oriented Environments

In service-oriented applications, effective and efficient trust evaluation is crucial. It provides valuable information to service clients, enabling them to select trustworthy service providers, consume high quality services and prevent monetary loss. One aspect of the work presented in this thesis is trust vector based approaches to trust rating aggregations in service-oriented environments. In the single trust vector approach, a trust vector of three values is computed, which can predict the trust trend and represent a set of ratings distributed within a time interval (e.g. a week or a month etc.). With trust vectors, two service providers with similar final trust values can be compared. In the multiple trust vector approach, a two dimensional aggregation is performed, which consists of both vertical and horizontal aggregations of trust ratings. The vertical aggregation calculates the aggregated rating representing the trust level for the services delivered in a small time period. The horizontal aggregation applies our proposed multiple time interval (MTI) greedy and optimal algorithms to determine the minimal number of time intervals, within each of which a trust vector can be calculated to represent all the ratings in that time interval. The proposed algorithms can return a small set of MTI to represent a large set of trust ratings and preserve the trust features well. The other aspect of the work presented in this thesis is trust-oriented composite service selection and discovery. In service-oriented environments, a service may invoke other services offered by different providers forming composite services. The complex invocations in composite services greatly increase the complexity of trustoriented service selection and discovery. To evaluate the subjective global trust of a composite service based on subjective probability theory, a set of subjective trust evaluation approaches have been proposed, which can map the non-binary discrete subjecv