Dynamic In-Order Parser

This repository includes the code of the in-order parser trained with a dynamic oracle described in EMNLP paper Dynamic Oracles for Top-Down and In-Order Shift-Reduce Constituent Parsing. The implementation is based on the in-order parser (https://github.com/LeonCrashCode/InOrderParser) and reuses part of its code, including data preparation and evaluating scripts.

This implementation requires the cnn library and you can find pretrained word embeddings for English and Chinese in https://github.com/LeonCrashCode/InOrderParser.

Building

The boost version is 1.5.4.

mkdir build
cd build
cmake .. -DEIGEN3_INCLUDE_DIR=/path/to/eigen
make

Experiments

Data

You could use the scripts to convert the format of training, development and test data, respectively.

python ./scripts/get_oracle.py [en|ch] [training data in bracketed format] [training data in bracketed format] > [training oracle]
python ./scripts/get_oracle.py [en|ch] [training data in bracketed format] [development data in bracketed format] > [development oracle]   
python ./scripts/get_oracle.py [en|ch] [training data in bracketed format] [test data in bracketed format] > [test oracle]

Training

mkdir model/
./build/impl/Kparser --cnn-mem 1700 --training_data [training oracle] --dev_data [development oracle] --bracketing_dev_data [development data in bracketed format] -P -t --pretrained_dim 100 -w [pretrained word embeddings] --lstm_input_dim 128 --hidden_dim 128 -D 0.2

Test

./build/impl/Kparser --cnn-mem 1700 --training_data [training oracle] --test_data [test oracle] --bracketing_dev_data [test data in bracketed format] -P --pretrained_dim 100 -w [pretrained word embeddings] --lstm_input_dim 128 --hidden_dim 128 -m [model file]

The automatically generated file test.eval is the result file.

For more information, please visit https://github.com/LeonCrashCode/InOrderParser.

Citation

@inproceedings{fernandez-gonzalez-gomez-rodriguez-2018-dynamic,
    title = "Dynamic Oracles for Top-Down and In-Order Shift-Reduce Constituent Parsing",
    author = "Fern{\'a}ndez-Gonz{\'a}lez, Daniel  and G{\'o}mez-Rodr{\'\i}guez, Carlos",
    booktitle = "Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing",
    month = oct # "-" # nov,
    year = "2018",
    address = "Brussels, Belgium",
    publisher = "Association for Computational Linguistics",
    url = "https://www.aclweb.org/anthology/D18-1161",
    doi = "10.18653/v1/D18-1161",
    pages = "1303--1313"    
}

Acknowledgments

This work has received funding from the European Research Council (ERC), under the European Union's Horizon 2020 research and innovation programme (FASTPARSE, grant agreement No 714150), from MINECO (FFI2014-51978-C2-2-R, TIN2017-85160-C2-1-R) and from Xunta de Galicia (ED431B 2017/01).