Ontology-based Natural Language Processing for In-store Shopping Situations

Natural Language communication between customers and products within in-store shopping environments enables new forms of product interfaces and an improved filtering and intuitive presentation of product information. In this article, we describe how customer's access to product information at the point of sale can be improved through the use of dialogue systems and heterogeneous web-based representations of product information based on formal ontologies within in-store shopping environments. After considering specific requirements of in-store shopping environments on dialogue systems, we present the model of a Conversational Recommendation Agent (CoRA), a domain-specific dialogue system, which realizes an ontology-based Natural Language Processing system for shopping situations.