Fig. 2From: Assisted neuroscience knowledge extraction via machine learning applied to neural reconstruction metadata on NeuroMorpho.OrgSchematic representation of the data for the metadata extraction task. Unstructured text of neuroscience articles are mapped to their corresponding metadata matching the NeuroMorpho.Org schema. Each article has a body of text extracted from PubMed or other publishers and is associated with a list of target metadata entities. The task of the algorithm is to find the target entities in the text of the article and create a list of suggestions ranked based on their relevancyBack to article page