Neural Relation Extraction（NRE）旨在将深度神经网络用于从文本抽取实体间关系构建知识图谱，清华大学刘知远老师课题组的林衍凯在去年ACL提出基于Sentence-Level Attention的CNN/PCNN技术，在标准测试集得到了远程监督关系抽取的最好效果。最近，清华大学刘知远老师课题组的本科生唐路明同学基于TensorFlow实现了该技术的LSTM版本，效果得到进一步显著提升，现将代码开源欢迎试用。

Neural Relation Extraction(NRE)

Neural relation extraction aims to extract relations from plain text with neural models, which has been the state-of-the-art methods for relation extraction. In this project, we provide our implementations of a word-level and sentence-level combined Bidirectional GRU network (BGRU+2ATT).

We come up with the idea from the paper "Attention-Based Bidirectional Long Short-Term Memory Networks for Relation Classification" [Zhou et al.,2016] and the paper "Neural Relation Extraction with Selective Attention over Instances" [Lin et al.,2016]. And we compare our results with PCNN+ATT [Lin et al.,2016] on the same dataset.

Evaluation Results

P@N comparison between PCNN+ATT and our method (BGRU+2ATT):

Precision/Recall curve of our method (BGRU+2ATT) compared to others':

链接：

https://github.com/thunlp/TensorFlow-NRE

原文链接：

http://weibo.com/1464484735/ExFiHxO9a?type=comment#_rnd1488371404201