Biomedical texts are a rich source of information. For instance, scientific publications such as MEDLINE abstracts disseminate the latest research results, and articles in online health portals provide the layman with information about diseases and therapies. All this information, however, is buried under the unstructured text such that it must be extracted via techniques such as text mining.
In a typical biomedical text mining pipeline, entity recognition and entity disambiguation are early tasks that enable further processing.
BEBE is a software that targets these two early tasks.
The distinguishing feature of BEBE is its speed, without sacrificing too much quality.
Amy Siu (Email: siu at mpi hyphen inf dot mpg dot de)