Fine-grained tourism prediction: Impact of social and environmental features

•Exploiting Social Media and Environmental features to perform accurate fine-grained (attraction-level) predictions.•Accurate prediction results (MAPE < 25%) for more than 93% of the indoor and outdoor attractions.•Collection, join and analysis of 5 different datasets containing two types of attractions - indoor and outdoor.•Better performance of Environmental features in outdoor attractions such as parks versus indoors like museums and galleries.•Complete Failure analysis for attractions in which proposed prediction methodology could not produce satisfactory results.