Skip to content

Latest commit

 

History

History
8 lines (7 loc) · 572 Bytes

File metadata and controls

8 lines (7 loc) · 572 Bytes

Sentiment-analysis-with-deep-neural-nets

The data folder contains negative and positive reviews from ecommerce platform.
The process_data folder contains train data set, test set and word dictionary. They are the results from data pretreatment.
The model folder contains deep neural network, which is consisted of Bi-directional GRU layer and attention mechanism.
The result folder contains prediction results from trained model.

The command to run the code is
python main.py -positive_reviews Data/pos.txt -negative_reviews Data/neg.txt