Integrating DBpedia and SentiWordNet for a tourism recommender system

The popularity of the social web introduces opportunities for the recommender systems, whilst new challenges arise when semantic knowledge is integrated in the landscape. The large amount of opinions available from Web 2.0 are exploited here to improve recommendation techniques in a semantic context. The developed recommendation system matches the crawled opinions against tourist objectives within the DBpedia ontology. Following a natural language processing step in Gate, several metrics are employed to build a recommendation plan, and formal justification is provided in case of need.