Stanford Named Entity Recognizer download

CRFClassifier is a Java implementation of a Named Entity Recognizer. The software provides an implementation of Conditional Random Field sequence models, of the sort pioneered by Lafferty, McCallum, and Pereira (2001), coupled with well-engineered feature extractors for Named Entity Recognition. Included are a good 3 class (PERSON, ORGANIZATION, LOCATION) recognizer (in versions with and without additional distributional similarity features) and another pair of models trained on the CoNLL 2003 English training data. The distributional similarity features improve performance but the models require considerably more memory.

The CRF code is by Jenny Finkel. The feature extractors are by Dan Klein, Christopher Manning, and Jenny Finkel. The CRF sequence model provided here does not precisely correspond to any published paper. The best paper to cite is:

Jenny Rose Finkel, Trond Grenager, and Christopher Manning. 2005. Incorporating Non-local Information into Information Extraction Systems by Gibbs Sampling. Proceedings of the 43nd Annual Meeting of the Association for Computational Linguistics (ACL 2005), pp. 363-370. http://nlp.stanford.edu/~manning/papers/gibbscrf3.pdf

The software provided here is similar to the baseline local+Viterbi model in that paper, but adds new distributional similarity based features (in the -distSim classifiers). You can look at a Powerpoint Introduction to NER and the Stanford NER package.

CRFClassifier requires Java 1.5. This distribution is licensed under the (full) GPL. Source is included.

The download is a 62 MB gzipped tar file (mainly consisting of classifier data objects). If you unpack that file, you should have everything needed. It includes batch files for running under Windows or Unix/Linux/MacOSX and the ability to run as a server. There is a list of Frequently Asked Questions (with answers!). Send any other questions or feedback to java-nlp-support@lists.stanford.edu.

Download Stanford Named Entity Recognizer version 1.5

Release history

Version 1.5 2008-05-07 Additional feature flags, various code updates
Version 1.0 2006-09-18 Initial release