Home

Awesome

DOER

The implementation of

DOER: Dual Cross-Shared RNN for Aspect Term-Polarity Co-Extraction. Huaishao Luo, Tianrui Li, Bing Liu, Junbo Zhang. ACL, 2019.

This paper focuses on two related subtasks of aspect-based sentiment analysis, namely aspect term extraction (ATE) and aspect sentiment classification (ASC), which we call aspect term-polarity co-extraction.

Requirements

pip install -r requirements.txt

Running

preprocess
python main.py --do_preprocess --data_sets laptops_2014
train
python main.py \
    --do_train --do_evaluate \
    --data_sets laptops_2014 \
    --choice_rnncell regu \
    --use_mpqa \
    --use_labels_length \
    --do_cross_share --lr 0.001 \
    --batch_size 16

See main.py for more training arguments.

word embedding

Amazon Embedding can be found in amazon, and Yelp Embedding can be found in yelp.

Citation

If this work is helpful, please cite as:

@Inproceedings{Luo2019doer,
    author = {Huaishao Luo and Tianrui Li and Bing Liu and Junbo Zhang},
    title = {DOER: Dual Cross-Shared RNN for Aspect Term-Polarity Co-Extraction},
    booktitle = {ACL},
    year = {2019}
}