Service discovery through consensus-based preferences

The prevailing research trend is to discover and compose web services dynamically in order to develop composite services that provide enhanced functionality. Most existing discovery techniques do not take into account the diverse preferences and expectations of service consumers. For instance, searching for a web service (for cheap and convenient flights), does not involve only functional requirements, but it also includes subjective and fuzzy opinions in query terms such as "cheap and convenient". Different perceptions of such terms by service consumers and providers create serious problems in service discovery. This is compounded by the fact that they may have different preferences or opinions. This paper presents a new service discovery approach to model subjective and fuzzy opinions, and to assist them in reaching a consensus. The method achieves a common consensus on the distinct opinions and expectations of individual users. This process is iterative and additional fuzzy preferences can be included to improve precision rate in web service discovery. © 2006 CRL Publishing Ltd.