summarization
所属分类:特征抽取
开发工具:Jupyter Notebook
文件大小:151KB
下载次数:0
上传日期:2018-05-03 15:43:32
上 传 者:
sh-1993
说明: 基于Keras的神经网络新闻文章文本摘要
(Text Summarization of news articles in Neural Networks using Keras)
文件列表:
summarization.ipynb (1849219, 2018-05-03)
# summarization
Text Summarization of news articles in Neural Networks using Keras
This project was undertaken to perform abstractive summarization on a news corpus by making use of deep learning models like encoder-decoder networks.
Tasks performed:
1. Developing a unidirectional LSTM based encoder-decoder network for title generation corresponding to each lede.
2. Improvement of the model using attention.
3. The scores were evaluated using ROUGE scores.
References:
1. http://colah.github.io/posts/2015-08-Understanding-LSTMs/
2. http://blog.echen.me/2017/05/30/exploring-lstms/
3. https://deeplearning4j.org/lstm.html
4. http://u.cs.biu.ac.il/~yogo/nnlp.pdf
近期下载者:
相关文件:
收藏者: