Gimli is an open source, high-performance and multi-corpus solution for automatic recognition of biomedical names. It uses Conditional Random Fields (CRFs) with a rich set of features, applying a method to combine different CRF models. Gimli achieves an F-measure of 87,54 on GENETAG and 73,05 on JNLPBA corpus, significantly outperforming existing open and closed source solutions.
- Linux/ Windows/MacOsX
:: MORE INFORMATION
Gimli: open source and high-performance biomedical name recognition.
Campos D, Matos S, Oliveira JL.
BMC Bioinformatics. 2013 Feb 15;14(1):54.