In recent years, several methods for personalizing information offerings have been developed to combat the information overload caused by the vast range and continued growth of content available on the World Wide Web. Examples include adaptive hypertext, which adapts the user interface to the users’ interests and needs (e.g., in mobile portals), and recommender systems that try to predict what items, e.g., news, movies or travels, would be relevant for the user. Recommender systems use various AI and IR techniques to build their predictions and typically interact with the user. Many of these techniques and systems have made their way from research into commercial applications and are now widely used in major eCommerce portals. At the same time, new techniques are being proposed, for improving the prediction accuracy or offering new ways for users to participate, as in social networks in Web 2.0 platforms.

Nowadays, different AI approaches are well-established as central technologies in many industrial configuration systems. This wide-spread industrial use of AI-based configurators makes the field more challenging than ever: the complexity of configurable products still increases, the mass-customization paradigm is extended to fields like service and software configuration, personalized (web-based) user interaction and user preference elicitation are of increasing importance, and finally, the integration of configurators into surrounding IT infrastructures like business information systems or web applications becomes a critical issue.

The main goal of the workshop is to promote high-quality research in all technical areas related to configuration. As such, the workshop is of interest for researchers working in the various fields within the wide range of applicable AI technologies (e.g. Constraint Programming, Description Logics, Non-monotonic Reasoning, Case-Based Reasoning ...). It serves as a platform for researchers and industrial participants to exchange needs, ideas, benchmarks, use cases etc. Collocated with ECAI 2008 the Workshop on Configuration Systems provides an ideal forum to attract high-quality submissions.

The workshop continues the series of seven successful Configuration Workshops started at the AAAI96 Fall Symposium and continued on IJCAI, AAAI, and ECAI since 1999. Beside the participation of researchers from a variety of different fields, past events always attracted significant industrial interest from major configurator vendors like ILOG, Oracle, or SAP, as well as from end-users like ABB, DaimlerChrysler, HP, or Siemens.

Driven by the strong industrial interest, the proposed workshop shall include an invited talk on current configuration technology, problem-solving approaches, and basic terminology before the technical sessions in order to provide a solid common ground for discussions and exchange of ideas between academia and industrial participants.

Furthermore, we invite tool vendors to submit position statements describing current problems, new challenges, experiences, or new application fields and use cases, that will be presented and discussed in a separate session and panel.

Finally, in order to attract high-quality technical papers, we will publish revised versions of selected papers in a special issue on configuration in "International Journal of Mass Customization". We may ask authors of journal submissions dealing with the same topic to write a joint article.

Recommender systems support information search and decision processes for the identification of (complex) products and services in e-Commerce environments as well as in other application areas such as digital libraries or e-learning portals. Recommendation problems have a long history as a successful application area of Artificial Intelligence and the interest in such systems has dramatically increased due to the demand for personalization technologies by large and successful e-Commerce platforms. The wide-spread industrial use of recommender applications makes the research field more challenging than ever. AI approaches to solving recommendation problems can be differentiated into Content-based, Collaborative, Knowledge-based, Demographic, and Utility-based ones. These approaches are based on methodologies overlapping other disciplines such as Human Computer Interaction, Natural Language Processing, Psychology or Sociology which makes Recommender Systems a strongly interdisciplinary research field.

The main goal of the ECAI’08 Workshop on Recommender Systems is to promote high-quality research in technical areas related to Recommender Systems. In order to fully exploit the potentials of recommender technologies, a strong focus has to be posed on the investigation of the human factor and the applicability of recommender technologies in real-world settings. Existing technologies have to be further developed in order to make recommender applications easier to apply and flexible to the requirements of users and business processes of suppliers. A number of related topics emerged in the last years which make the workshop interesting for researchers and participants from industry alike.

The main goal of the ECAI’06 Workshop on Recommender Systems is to promote high-quality research in technical areas related to Recommender Systems. In order to fully exploit the potentials of recommender technologies, a strong focus has to be posed on the investigation of the human factor and the applicability of recommender technologies in real-world settings. Existing technologies have to be further developed in order to make recommender applications easier to apply and flexible to the requirements of users.

Extended versions of selected papers will be published in a special issue on "Recommender Systems" in IEEE Intelligent Systems.