||
一、自动文本摘要
(1)Sequence-to-Sequence with Attention Model for Text Summarization:
https://github.com/tensorflow/models/tree/master/research/textsum
论文:Rush et al. A Neural Attention Modelfor Sentence Summarization.
(2)Preparing a dataset for TensorFlow text summarization (TextSum) model:(数据处理的参考)
https://github.com/surmenok/TextSum
(3)传统文本摘要,抽取式的
https://github.com/zkwi/textSummary
(4)Uses Recurrent Neural Network (LSTM and GRU units) for developing Seq2Seq Encoder Decoded model with and without attention mechanism for summarization of amazon food reviews into abstractive tips:
https://github.com/harpribot/deep-summarization
(5)fasttext和sentence_extractor代码参考:
https://github.com/yangze01/NeuralSummarization/tree/master/model
(6)tl-dr (abstractive text summarization):GRU encoder-decoder model and a QRNNenc + RNNdec model.
https://github.com/padelson/tl-dr2
(7)Abstractive-Summarization-using-Query-based-Deep-Neural-Attention-Models
(8)A neural news summarization tool. Collect google news topics, crawl related news articles, generate summaries.Including the following units:
BasicSum: traditional freq-based news-summary generator.
LSTM-Attetion: Neural network model that summarizes news.
WikiNews Dataset: A list of news articles crawled from WikiNews, every element(event) of the list contains at least three articles from mainstream website talking about the event.
https://github.com/qinenergy/NewsSum
(9) Sequence to sequence model for abstractive text summarization
https://github.com/weichengzhang/Summarization
(10)
二、seq2seq学习
(1)Sequence to sequence (seq2seq) learning Using TensorFlow.
https://github.com/JayParks/tf-seq2seq
(2)Neural Sequence Learning Using TensorFlow
https://github.com/ufal/neuralmonkey
(3)生成古诗
https://github.com/Disiok/poetry-seq2seq
(4)
三、机器阅读理解
(1)Bi-directional Attention Flow for Machine Comprehension
https://github.com/allenai/bi-att-flow
四、数据可视化
(1)A tool for finding distinguishing terms in small-to-medium sizedcorpora, and presenting them in a sexy, interactive scatter plot withnon-overlapping term labels. Exploratory data analysis just got more fun.
https://github.com/JasonKessler/scattertext
Archiver|手机版|科学网 ( 京ICP备07017567号-12 )
GMT+8, 2024-5-18 04:46
Powered by ScienceNet.cn
Copyright © 2007- 中国科学报社