Exploiting Codified User Task Knowledge to Discover Services at Design-Time

Most techniques for engineering service-based applications do not explicitly exploit knowledge about users and their tasks. In this paper the authors hypothesize that codified knowledge about user tasks can improve service discovery at design-time. It reports the extension of an existing service discovery algorithm to match service queries to user task models then reformulate the service queries with task-specific knowledge in order to improve discovery precision and recall scores. An empirical investigation of the extended algorithm revealed that, in order to deliver significant benefits, user task models need to describe more context-specific knowledge with which to extend service queries.