Sustainable Tourism: Crowdsourced Data for Natural Scene and Tag Mining

Abstract

In recent years, the concept of sustainable tourism has emerged at the intersection of debates on visiting somewhere as a tourist and trying to make a positive impact on the environment, society, and economy. By leveraging the power of online infrastructures, we demonstrate that crowdsourced generated data, by the tourists, encode a vast amount of information, such as the physical properties from the photo and description from textual information. Using these online platforms, such as Flickr, users generate crowdsourced geotagged information containing an immense amount of human behavior tracking on scenic views. In this paper, geotagged Flickr data is used for automatic natural scenes classification using an image, and textual features obtained from the crowdsourced data. The proposed method uses the data mining technique with descriptors. The results show that the geotagged Flickr data can imply Urban City interaction with an encouraging accuracy of 90.20% and that the proposed approach improves natural scene classification efficiency if a sufficient spatial distribution of crowdsourced data exists. Hence, social sensing mining the attractiveness of human interaction is an interesting or tourism area using image processing and text mining method with geotagged mobility of users can provide accurate information that challenges for developing sustainable tourism management.