Example of a moving window of n = 6 of a skip-n-gram model over a piece of text from O'Farrill et al. (2013). The text has been cleaned of common stop words (e.g. "the", "all", "however"). Inside the moving window, a central word is fixated (randomly) and all possible word pairs are considered as word vectors. After this step is completed, the moving window advances one word and repeats the process again. Frequencies of co-occurrences within the pool of word vectors are further used to rank word pairs.

  Part of: Muñoz G, Kissling W, van Loon E (2019) Biodiversity Observations Miner: A web application to unlock primary biodiversity data from published literature. Biodiversity Data Journal 7: e28737. https://doi.org/10.3897/BDJ.7.e28737