Recommending prime spots of a destination and time to visit from geo-tagged social data

Abstract

Planning a trip can be a tedious task. One has to search for what places to visit at a destination (i.e. area) and what time to visit the destination. Sometimes this can be a time-consuming task because there are too much information available, and it is hard for one to choose which information to trust. In this paper we present a recommendation system clustering geo-tagged social data in a destination from each information source - Flickr and Foursquare - and combining the results from these diverse information sources to recommend places to visit. Our experimental results show that our recommendation system automatically suggests prime spots in Yellowstone national park with 0.83 precision and 0.927 NDCG, and in Yosemite national park with 0.8 precision and 0.912 NDCG. In addition, visualizing temporal information of social data helps travelers to decide when to visit a destination.