We consider an approach to service selection wherein service consumers choose services with desired nonfunctional properties to maximize their utility. A consumer’s utility from using a service clearly depends upon the qualities offered by the service. Many existing service selection approaches support agents estimating trustworthiness of services based on their quality of service. However, existing approaches do not emphasize the relationship between a consumer’s interests and the utility the consumer draws from a service. Further, they do not properly support consumers being able to compose services with desired quality (and utility) profiles.
We propose an adaptive service selection framework that offers three major benefits. First, our approach enables consumers to select services based on their individual utility functions, which reflect their preferences, and learn the providers’ quality distributions. Second, our approach guides consumers to construct service compositions that satisfy their quality requirements. Third, an extension of our approach with contracts approximates Pareto optimality without the use of a market mechanism.