A Tensorflow implementation for "Recurrent Attention Network on Memory for Aspect Sentiment Analysis" (Peng Chen, EMNLP 2017)
- Create three empty folders: 'analysis' for saving analyzing results, 'logs' for saving experiment logs and 'models' for saving experiment models
- Download the 300-dimensional pre-trained word vectors from Glove and save it in the 'data' folder as 'data/glove.840B.300d.txt'
|--- data
| |--- restaurant
|--- main.py
|--- model.py
|--- utils.py
|--- README.md
Dataset | Accuracy |
---|---|
Laptop | 74.138 |
Restaurant | 80.000 |