Multiple annotation for biodiversity: developing an annotation framework among biology, linguistics and text technology

Abstract Biodiversity information is contained in countless digitized and unprocessed scholarly texts. Although automated extraction of these data has been gaining momentum for years, there are still innumerable text sources that poorly accessible require a more advanced range methods to extract relevant information. To improve the access semantic biodiversity information, we have launched BIOfid project ( www.biofid.de ) developed portal semantics German language texts, mainly from 19th 20th century. However, make such work, couple had be or adapted first. In particular, text-technological were needed, which required Such draw on machine learning techniques, turn trained by data. this end, among others, gathered bio corpus, cooperatively built resource, biologists, technologists, linguists. A special feature its multiple annotation approach, takes into account both general biology-specific classifications, means goes beyond previous, typically taxon- ontology-driven proper name detection. We describe design decisions genuine Annotation Hub Framework underlying annotations present agreement results. The tools used create introduced, use described. Finally, some lessons, particular with projects, drawn.