Automatic generation of natural language service descriptions from OWL-S service descriptions

• Autor(a) principal: Gonçalves, Luís Filipe Delgado Alves Ribeiro
• Data de Publicação: 2009
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/23213

As the web grows in both size and diversity, there is an increased need to automate aspects of its use such as service coordination (e.g., discovery, composition and execution). Semantic web services combine semantic web and web service technologies, providing the support for automatic service coordination. Semantic web services are described using semantic languages (e.g., OWL-S) and can be automatically processed by intelligent agents (agent based coordination). This dissertation aims at enhancing the service coordination process, building upon well-understood and widespread practices on natural language generation. Automated service coordination relies on the existence of formal service descriptions (semantic languages, such as OWL-S or WSML). The use of web services by people is essentially associated with the discovery, composition and execution of services that match their needs. According to the person’s will, the discovered or composed service is or is not executed. This decision can only be made if the person understands the description of the service. Therefore, it is necessary that formal descriptions be converted into more natural descriptions, adequate to human comprehension. This dissertation contributes to empower the users (knowledge engineers and common citizens) of service coordination systems with the capability to better understand and decide about discovered or composed services without the need of understanding the formal language in which the semantic web service is described. We implemented a software program capable of generating natural language service descriptions from OWL-S description. It is a template-based natural language generation system that receives the OWL-S description of a service as input and converts it into an English description. This system will leverage the use of service coordination technology by people and allow them to have a more active role in the various stages of the service coordination process.