Adaptive Multi-Task Transfer Learning for Chinese Word Segmentation in Medical Text

Code & Data for our COLING 2018 paper "Adaptive Multi-Task Transfer Learning for Chinese Word Segmentation in Medical Text"

Processed open-source datasets are available in data folder. Annotated medical forum dataset is available in medical_data folder.

Run python3 main.py -h to see how to specify your own model settings or datasets.

For training, use the command python3 main.py train -t 2 -p mmd to train the model with model type 2 and mmd penalty.

Contact

Pls drop me (Junjie Xing) an email at gavinxing9016 at gmail.com if you have any question.