Automation of Web Service Provisioning Process using Contextual Knowledge of Users

Area description

Web Services enhanced with semantics, known as Semantic Web Services, have been in an active development for one decade now. They leverage expressive semantic languages for description of services and they use logical reasoning and problem-solving methods to enable automation of service provisioning (such as discovery, selection, composition, mediation and invocation of services). Many works in this area have built problem-solving infrastructures that act on behalf of users with one goal to reduce their manual activities and replace them by intelligent behavior of software agents. The major results in this area include semantic specifications for Web services such as Web Service Modeling Ontology (WSMO), WSMO-Lite, OWL-S, as well as Semantic Annotations for WSDL and XML Schema (SAWSDL). On top of these specifications, there exist a variety of works defining algorithms for service discovery, composition, selection and mediation.

Problem Statement

Although Semantic Web Services provide very sophisticated automation methods, they usually assume that software agents can ultimately replace users' activities. They rely only on an explicit desired state and a desired quality that should result from the service provisioning and not how users could participate in the provisioning process. They do not take into account a wider user's environmental context such as his/her presence in social media, social relationships with other users, relationships among multiple users' goals, and activities that users perform (observable user behavior) within the service provisioning. Users' collective experience and users' social context is a very essential kind of knowledge that can complement sophisticated automation methods and can significantly improve the results of service provisioning.

Tasks and Goals

Following table shows types of activities from the methodology and expected research work in that activity. Please read the research guidelines to understand the context.

Methodology Tasks and Goals
Analysis/Design Augmenting service descriptions with social ontologies using microformats, RDFa, SAWSDL or utilization of linked data for service descriptions.
Analysis/Design Recommendation algorithms for Web services utilizing collective and social knowledge and improving service discovery, selection, and composition through feedbacks and critics.
Implementation Implementation of a prototype for social services recommendation system.
Evaluation Evaluation of algorithms on data from Web service directories such as