Paper: Boosting the protein name recognition performance by bootstrapping on selected text

ACL ID W12-2430
Title Boosting the protein name recognition performance by bootstrapping on selected text
Venue Workshop on Biomedical Natural Language Processing
Session
Year 2012
Authors

When only a small amount of manually anno- tated data is available, application of a boot- strapping method is often considered to com- pensate for the lack of sufcient training ma- terial for a machine-learning method. The paper reports a series of experimental results of bootstrapping for protein name recogni- tion. The results show that the performance changes signicantly according to the choice of text collection where the training samples to bootstrap, and that an improvement can be obtained only with a well chosen text collec- tion.