This is the implementation of Summarizing Medical Conversations via Identifying Important Utterances at COLING 2020.

You can e-mail Yuanhe Tian at [email protected] or Guimin Chen at [email protected], if you have any questions.

If you use or extend our work, please cite our paper at COLING 2020.

@inproceedings{song-etal-2020-summarizing,
    title = "Summarizing Medical Conversations via Identifying Important Utterances",
    author = "Song, Yan and Tian, Yuanhe and Wang, Nan and Xia, Fei",
    booktitle = "Proceedings of the 28th International Conference on Computational Linguistics",
    month = dec,
    year = "2020",
    address = "Barcelona, Spain (Online)",
    pages = "717--729",
}

Our code works with the following environment.

• python=3.7
• pytorch=1.3

To obtain the data, you can go to data_preprocessing directory for details.

In our paper, we use BERT (paper) and ZEN (paper) as the encoder.

For BERT, please download pre-trained BERT-Base Chinese from Google or from HuggingFace. If you download it from Google, you need to convert the model from TensorFlow version to PyTorch version.

For ZEN, you can download the pre-trained model from here.

For HET-MC, you can download the models we trained in our experiments from here (passcode: b1w1).

Run run_sample.sh to train a model on the small sample data under the sample_data directory.

You can find the command lines to train and test models in run.sh.

Here are some important parameters:

• --do_train: train the model.
• --do_test: test the model.
• --use_bert: use BERT as token encoder.
• --use_zen: use ZEN as token encoder.
• --bert_model: the directory of pre-trained BERT/ZEN model.
• --use_memory: use memories.
• --utterance_encoder: the utterance encoder to be used (should be one of none, LSTM, and biLSTM).
• --lstm_hidden_size: the size of hidden state in the LSTM/biLSTM utterance encoder.
• --decoder: the decoder to be used (can be either crf or softmax).
• --use_party: use the speaker role information.
• --use_department: use the department information.
• --use_disease: use disease information
• --model_name: the name of model to save.

• Release the code to get the data.
• Regular maintenance.

You can leave comments in the Issues section, if you want us to implement any functions.