MEMOS: A Methodology for Modeling Services

                                  Mick Kerrigan

                             Supervisor: Elena Simperl



Abstract
Research in the area of Semantic Web Services aims to totally or partially auto-
mate activities that occur in the process of using Web services. Such automation
is required due to the large amounts of human effort required to find, select, un-
derstand, test, and invoke Web services, which becomes infeasible as the number
of services on the Web steadily grows. Automation of activities like service dis-
covery, composition, selection, mediation, and invocation within a Semantic Ex-
ecution Environment is achieved through the creation of semantic descriptions
of the available services and of the requirements of a given service requester. It is
important that engineers on both the service provider and the service requester
sides have a clear understanding of the tasks they must perform to create these
descriptions. To date no methodological support has been available to these en-
gineers. Individual development projects have had to spend significant effort in
understanding what descriptions need to be created to support a particular use
case and how to go about creating these descriptions.
    The aim of this work is to assist engineers in the process of developing Seman-
tic Web Services, by providing a clear understanding of what description need
to be created in particular use cases, by specifying the different tasks that must
be performed in order to successfully create these descriptions, and by offering
tool support that reduces the effort needed to create the descriptions. The main
contribution of this work is a Methodology for Modeling Services (MEMOS)
that defines activities and tasks that should be conducted by different roles in
the software development cycle of Semantic Web Services. The methodology is
built around a collection of Semantic Web Service engineering scenarios, which
identify the different Semantic Web Service artifacts that need to be semanti-
cally described in order to enable each scenario. The individual scenarios can
be combined to create complex use cases, while still maintaining a clear under-
standing of exactly which descriptions need to be created to enable such a use
case. Engineers are assisted in the execution of individual tasks in the methodol-
ogy by the Web Service Modeling Toolkit (WSMT), an integration development
environment for Semantic Web Services that provides tool support for many of
the mundane or error prone tasks.