Author: Jiang Guo, [email protected]
This tool is used for semi-supervised learning of NER, using various kinds of embedding features. This tool is associated with (Guo et al., 2014). The proposed approaches are shown to be much better than the direct usage of continuous word embedding features.
The original dataset should be converted to BIO-style annotation.
$ ./train.sh [de|bi|ce|proto]
de
- dense embedding featuresbi
- binarized embedding featuresce
- clustered featuresproto
- distributional prototype features
To use the combined features, e.g. de+proto
$ ./tag.sh [de|bi|ce|proto]
@InProceedings{guo-EtAl:2014:EMNLP2014,
author = {Guo, Jiang and Che, Wanxiang and Wang, Haifeng and Liu, Ting},
title = {Revisiting Embedding Features for Simple Semi-supervised Learning},
booktitle = {Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
month = {October},
year = {2014},
address = {Doha, Qatar},
publisher = {Association for Computational Linguistics},
pages = {110--120},
url = {http://www.aclweb.org/anthology/D14-1012}
}