Semantic Web travel services on a voyage of discovery

Semantic Web travel services on a voyage of discovery

A secure, semantic-based interoperability framework for exploiting Web-service platforms across peer-to-peer networks in the tourism industry hopes to set free the valuable Web-based tourism information that is currently trapped in isolated silos of incompatible databases. "The tourism industry today is the second largest economic sector, after manufacturing, in the world," says Professor Asuman Dogac, Director of the Software Research & Development Center at the Middle East Technical University in Ankara, Turkey, and coordinator of the IST-funded SATINE project. "Earlier than in other sectors, tourism embarked on e-business, not only with respect to business-to-business, but also with business-to-consumer." Currently, travel information services are dominantly provided by Global Distribution Systems (GDSs). A GDS gives its subscribers pricing and availability information for multiple travel products such as flights. Travel agents, corporate travel departments, and even Internet travel services, subscribe to one or more GDS’. "However, small and medium-sized enterprises [SMEs], for example bed-and-breakfast accommodation or companies hiring bicycles, restaurants and many others cannot participate in GDS-based e-business because selling their products through GDS’ is too expensive for them," says Dr Dogac. Furthermore, GDSs are legacy systems, and suffer from a number of problems. They mostly rely on private networks, are mainly for human use, are difficult to use and have cryptic interfaces. Their speed and search capabilities are limited. Finally, it's difficult to get them to interoperate with other systems and data sources. This means that tour operators, travel agencies and others cannot benefit fully from the advantages of electronic business-to-business trading. Opening up travel Web services The industry is making moves to improve the situation, and has formed a consortium called the Open Travel Alliance (OTA). The OTA is producing XML message specification schemas to be exchanged between the trading partners, including availability checking, booking, rental, reservation, query services, insurance. This will improve matters, but not every travel company's applications can be expected to produce and consume OTA-compliant messages. The SATINE project developed a secure peer-to-peer network that enables peers to deploy their semantically-enriched travel Web services and allows others to discover these services semantically. "The concept, ‘semantic web services’, is very important for the tourism industry since the industry is structured as a distributed environment, because of its nature, and tourism companies need to reach the services in this distributed environment," says Dr Dogac. "By introducing semantics to Web services, we have addressed the interoperability issue on the semantic level and constructed a peer-to-peer network that eases service discovery." The project also developed mechanisms to enrich ebXML (e-business XML) registries through OWL-S ontologies to describe the Web service semantics. OWL itself is a markup language for publishing and sharing data using ontologies on the Web. Ontologies are the vocabularies that allow machines to identify specific services or information. For example, a human operator would understand the term 'booking', but it needs to be defined in a special way for a machine to understand. But getting the machine to correctly identify a 'booking' Web service is a difficult job and takes a lot of work. Luckily, Dr Dogac and her team are one of the leaders in the field. The upshot, though, is that once a machine can correctly identify concepts 'booking' or other travel-related terms, it not only allows for interoperability between incompatible systems, it also unlocks the potential for Web-based information services and could ultimately mean very complex operations are performed in minutes, instead of hours or days. The potential for advanced online travel Web services For instance, ultimately the work of SATINE could lead to very advanced travel services, where a holiday maker might enter terms like the dates, cost, car rental, hill walking, scuba diving and sunbathing and then be presented with a list of options offering all those activities and services with detailed itineraries and cost information, and all necessary bookings. The user would simply click on the option that appeals most. We're not there yet, but on the way to realising this dream for future travel information services, the SATINE project described how the various constructs of OWL can be mapped to ebXML classification hierarchies and how the services are discovered through standardised queries by using the ebXML query facility. This approach is for tourism companies or a group of such companies to use service registries efficiently for locating their Web services semantically to enable semantic searching. "The creation of complex services through the orchestration of simple Web services is an important task that is of particular relevance in the travel business: Apart from the typical examples, like the composition of package tours, more sophisticated services like a flight booking based on the availability of tickets for a certain cultural event are conceivable," says Dr Dogac. So the SATINE project is developing a semantic Web service composition and execution framework. "In SATINE project, we have developed a Semantic Wrapper for constructing and describing Web services," says Dr Dogac. The main role of this component is to wrap existing information resources to make them appear as semantically well-described Web Services. It's an easy-to-use tool for SMEs to create and annotate Web services from their existing enterprise applications. The Semantic Wrapper provides two complementary tools: the Web Service Creator and the Web Service Annotator. The Web Service Creator transforms existing resources in Web Services. The Web Service Annotator describes a Web service at a semantic level, using OWL-S as the ontology of reference. The project is also developing a mobile application to wrap the basic functionality of SATINE and provide a mobile interface for consumers to query and invoke the services in the network. These are secure Web services deployed among peers in the p2p network. The project developed a prototype that was demonstrated successfully at over six major conferences. SATINE also organised a number of commercial demonstrations, including one to German software firm, SAP research. AHEAD Relationships in Greece wants to exploit these technologies as part of its current business. Right now possible exploitation paths is under discussion within the SATINE consortium. Once Licensing Policy and the intellectual property rights sharing are finalised, the consortium will contact AHEAD Relationships to further discuss opportunities. Contact: Prof Dr Asuman Dogac Department of Computer Engineering Director of Software Research & Development Center Middle East Technical University 06531 Ankara Turkey Tel: +90-312-2105598 / 2102076 Fax: +90-312-2101004 / 2101259 Email: asuman@srdc.metu.edu.tr Source: Based on information from SATINE