This is a Tensorflow implement of the proposed in paper "Adversarial Learning of Sentiment Word Representations for Sentiment Analysis"

1. Python 3.6
2. Tensorflow 1.10 (<2.0)

Sentiment140 (Divided into two files, one contains positive sentences and the other contains negative sentences)

The training datasets and our results can be downloaded from https://pan.baidu.com/s/1gt9hZR6zAwoTWmsd2uaquw (Extraction code: 6nv6). (Github传大文件太慢了,在下懒了)

• Word-level: https://github.com/mfaruqui/eval-word-vectors

• Sentence-level: https://github.com/facebookresearch/SentEval

• https://www.tensorflow.org/versions/r0.10/tutorials/word2vec/index.html
• https://github.com/tensorflow/tensorflow/blob/r0.10/tensorflow/examples/tutorials/word2vec/word2vec_basic.py

  @article{BO2020,
    title = "Adversarial Learning of Sentiment Word Representations for Sentiment Analysis",
    journal = "Information Sciences",
    year = "2020",
    issn = "0020-0255",
    doi = "https://doi.org/10.1016/j.ins.2020.06.044",
    url = "http://www.sciencedirect.com/science/article/pii/S0020025520306320",
    author = "Peng Bo and Wang Jin and Zhang Xuejie",
    }