Near2me: an authentic and personalized social media-based recommender for travel destinations

This paper presents Near2me, a prototype system implementing a travel recommender concept that generates recommendations that are not only personalized, but also authentic. Exploitation of implicit situational knowledge makes it possible for Near2me to recommend places that are not necessarily touristic or famous, but rather are genuinely representative of place and also match users' personal interests. The system allows users to explore, evaluate, and understand recommendations, control recommendation direction and discover informative supporting material. This functionality makes it possible for users to assess recommendations and confirm their suitability and authentic nature. The recommendation system makes use of user photos from the image sharing community Flickr. We take the position that a social media-based environment incorporating multimedia content items, user-contributed annotations and social network connections is uniquely suited to providing users with authentic, personalized recommendations. First results of a user study allow us to conclude that users are interested in exploring locations, topics, and people from different perspectives and confirm authenticity as a relevance criterion.