Using social media photos to identify tourism preferences in smart tourism destination

Tourism and Travel sector are improving services through the use of a large amount of data gathered from different sources. The ease access to reviews, rating and experiences from different tourists made the touristic planning rich and complex. Hence, a big challenge faced by tourism sector is to use the gathered data for detecting tourist preferences. In this work, we proposed a solution able to detecting tourist implicit preferences based on photos. We used emerging techniques such as Convolutional Neural Network and fuzzy logic to classify tourists in five categories. Finally, a set of photos were analyzed and the results reached over 82% on accuracy metric for all classifiers.