Notes from Nature - How Weird is That?

[How Weird is That?] (+tab+https://biospex.org/projects/how-weird-is-that) classifies the language used by biodiversity research specimen collectors into that language expressing an unexpected observation from that which doesn’t. These specimens include plants on sheets, fish in jars, insects on pins, fossils in drawers, and more. Each specimen image seen by participants should include an accompanying label that uses a word that the collection community previously recognized as being used to document anomalies (e.g., “early”, “late”, “odd”, “weird”, “than”; Pearson and Mast, 2019, American Journal of Botany). The next step after this is to build machine learning algorithms to flag records likely to contain an observation of a biological anomaly to get this information to the researchers, conservation biologists, natural resource managers, and others who could use it. But to do this, we need to present the software with training data that has many examples of language that does and does not contain observations of anomalies. Building that training set is the goal of How Weird is That?