Project: Ontology-based Product Data Management

Internet use reached new records all over Europe in 2009. Currently more than 74% of the West-European population and more than 65% of the East-European population are already using the Web. As these figures have grown, e-commerce becomes more and more important. In Germany, for the first time, more than 50% of the total revenues of mail order companies were generated within the e-commerce segment. Referring to the TNS Infratest e-commerce study 2009, Germany is the global leading country in e-commerce with a per-capita e-commerce-spending of 913 EUR in 2008 (USA: 903 EUR) – Based on the given figures for Germany, which is the most developed e-commerce market in Europe, a huge potential for the growth of e-commerce markets all over Europe is expected within the next 5 to 15 years (as for Germany, a further significant increase is expected, e.g. 1.444 EUR per capita in 2012). In fact, there is a yearly shift of about 1%-5% of overall revenues (depending on the doCO) from walk-in stores to the online environment. Looking beyond Europe, the USA e-commerce market is quite well developed, while most emerging markets like China and India are currently at the beginning of e-commerce developments. Hence, despite the perception that the e-commerce market is already well-known and/or saturated, it still offers a high potential for continuing growth and development in the European, as well as in the global context._x000D__x000D_While many technologies and strategies within the realm of e-commerce applications depend on their individual doCO, there is one significant topic that is common to all kinds of approaches, and is one of the major challenges for current applications and future developments: e-commerce Application Data. It encompasses varied data required by e-commerce applications, such as product details, product contents and meta-information. Data has been collected and merged inefficiently over many years without considering data structuring and intelligent data management. This leads to bottlenecks within the workflows between product data, intelligent shop systems and e-commerce applications, which could be averted, if product data and information was represented in a structured way. Today, many market players that are active as mail order companies, manufacturers, distributors, etc. are looking for applications and business processes to solve the present problems in data management. They intend to tackle the issue of bottlenecks, since they do understand that improvements in this area could lead to a significantly better performance of their e-commerce applications. The described problem and the development of applicable and qualified solutions to solve it are the CO focus and motivation of the OPDM-project._x000D__x000D_The OPDM industrial consortium Ps have been researching solutions to address the issue of intelligent product and doCO data management in e-commerce applications. They came to the conclusion that a product data management system, which is based on Semantic Web-Technologies, is the most sensible way to approach this topic. Moreover, a product data management system is highly beneficial to complex e-commerce applications and can lead to a global technology leadership within the related market areas. _x000D__x000D_The CO objective of the project is to develop an application that can be integrated into existing shop systems, product information management systems, and third party applications and is able to handle product data based on individual doCO ontologies. It should provide components to reduce the required resources for manual data transformation and manual data management to a potential minimum. The backend of the application will focus on semantic data transformation heuristics and algorithms, while the frontend will concentrate on the graphical user interactions to, as a whole, provide an easy-to-handle web application that produces complex and highly useful doCO ontologies and associations on the data level. The project will be based upon two very successful approaches within the area of Semantic-Web-Technologies: The Linked Open Data (LOD) approach will be used and implemented within the project in order to include existing structured data. The GoodRelations-Ontology, which is already a global standard for describing offerings within the Semantic Web, will be used as a basic model for data representation in the application. _x000D__x000D_The consortium consists of leading e-commerce SME, such as the Austrian Semantic-Web-company Smart Information Systems, the German shop system provider Intershop and TIE from the Netherlands, which has been dealing with complex data transformation scenarios in several research projects in the past. To handle the difficult challenge of well-designed user-interfaces for complex ontology-engineering processes, UDE will cover this topic. UNIBW, as a leading organization for e-commerce ontologies, will develop the underlying data model of the application. _x000D_

Acronym OPDM (Reference Number: 5626)
Duration 01/01/2011 - 31/12/2013
Project Topic E-commerce is reaching new records worldwide. While e-commerce applications in general are differentiated, most of them face technical difficulties regarding data management. OPDM will provide an innovative application to handle e-commerce sufficiently.
Project Results
(after finalisation)
Over many years, product data has been collected and merged inefficiently without considering data structuring and intelligent data management. On the long run, this will cause, and already causes, bottlenecks in the workflows between data handling, intelligent shop systems and e-commerce applications, which could be averted, if product data and information was represented in a structured and homogeneous form. OPDM aimed at providing a solution, based on Semantic Web technologies, that supports manufacturers, content providers and retailers manage and process their product data more efficiently. The CO outcomes of the project are: A) OPDM webbased Ontology and Datamanagement Platform with interfaces to existing shopsystems A web-based solution with an easy-to-use interface to create and manage product vocabularies (product ontologies) and product data. In contrast to solutions available on the market, OPDM is particularly suitable for users without special knowledge of Semantic Web technologies. The application provides users with the following functions: - Create, browse and upload product vocabulary or clone existing vocabularies - Create and manage different product vocabulary profiles - Browse, upload and monitor product data - Import and export of structured product data from and to into third-party applications B) Semantic Data Transformation - TIE Semantic Integrator The application is able to analyze semi-structured CSV and XML-formatted product data, to automatically map identified product attributes to existing product vocabularies (Automated Mapping) - and thereby transform and structure product data according to existing schema (Data Transformation). C) 50 Product vocabularies based on the GoodRelations Ontology meta model To provide future users with direct access to existing product vocabularies, which they can use as a starting point, we developed product vocabularies (ontology templates) with detailed specifications and comprehensive descriptions in 50 popular e-commerce categories. The product vocabularies (ontologies) are compatible with the GoodRelations ontology - a widely used and standardized vocabulary for product, price, store, and company data. The developed product vocabularies are publicly available and can be obtained from to spur and stimulate innovations - both in scientific as well as commercial contexts. D) Validation of the project results within real-world case studies The evaluation of the OPDM results in terms of market relevance and real-world applicability was carried out within two case studies. - Development of 20 online product advisors using OPDM-structured product data - Implementation of OPDM-interfaces to the Intershop shop system
Network Eurostars
Call Eurostars Cut-Off 4

Project partner

Number Name Role Country
5 Intershop Communications AG Partner Germany
5 University of Duisburg-Essen Partner Germany
5 Universität der Bundeswehr München Partner Germany
5 TIE Holding N.V. Partner Netherlands
5 Smart Information Systems GmbH Coordinator Austria