An Accommodation Recommender System based on Associative Networks

In this paper we present a natural language interface that allows for easy and intuitive access to tourism information. In particular, we describe the knowledge representation model underlying the information retrieval system based on associative networks and that allows the definition of semantic relationships between domain-intrinsic information items. The network structure is used to define weighted associations between information items and we show how the system is enriched by a fuzzy search strategy. A constrained spreading activation algorithm implements information retrieval on the associative network. Strictly speaking, we take the relatedness of information items into account and show how this search strategy yields results highly associated to users’ queries. Furthermore, determining appropriate associations between information items is crucial. Thus, we propose an approach based on past user interactions for identifying semantic relations between information items.