Predicting environmental features by learning spatiotemporal embeddings from social media

Shelan S. Jeawak, Christopher B. Jones, Steven Schockaert

Abstract: Highlights•We use social media as a supplementary source of ecological information.•We propose a novel spatiotemporal embeddings model (SPATE) that integrate Flickr tags with structured scientific information.•We also propose a method to handle the sparsity of Flickr tag distributions over space and time.•The SPATE model is used to predict a wide variety of environmental features such as species distribution and climate.