ECAI 2004 Conference Paper

[full paper]

ADVISOR SUITE - A knowledge-based sales advisory system

Dietmar Jannach

In many domains, a large variety of comparable products and services from different manufacturers is available on the market. As a result, customers are increasingly overwhelmed when they have to choose the products that match their personal needs and preferences because the selection process in many cases requires deep technical knowledge about the product features. In the online sales channel, recommender systems have already been successfully applied as an additional means for increased customer guidance. Real-world sales advisory, however, can be a knowledge-intensive task. In order to elaborate the real customer needs, an experienced sales assistant will for instance conduct a dialogue on the customer's knowledge level, match these needs with the available products, and will finally explain the recommendations as well as possible alternatives. In this paper we present CwAdvisor, a now-commercialized system for building such intelligent, personalized sales advisory applications. CwAdvisor consistently follows a knowledge-based approach where all required development activities - from the definition of the user model and the recommendation rules, up to the personalization of the dialogue flow as well as user interface generation - are based on simple conceptual models and declarative knowledge representation. All these tasks are also supported by a set of graphical tools comprehensible for non-programmers, which leads to significant reductions in development and maintenance costs. At the end of the paper we discuss experiences from several commercial installations of the presented system.

Keywords: PAIS

Citation: Dietmar Jannach: ADVISOR SUITE - A knowledge-based sales advisory system. In R.López de Mántaras and L.Saitta (eds.): ECAI2004, Proceedings of the 16th European Conference on Artificial Intelligence, IOS Press, Amsterdam, 2004, pp.720-724.