CNN-RelationExtraction

Convolution neural network for relation classification between two given entities

The CNN architecture implemented is inspired be Nguyen et al. 2015 in which for each:

##Future work:

• Implementing additional crafted features to vectorizers/relationmentionvectorizer.py such as pos tags and dependency relations.
• Implementing sentence based features : in which additional sentence based features are added including two head words to find relations between them
  

References:

• Nguyen, Thien Huu, and Ralph Grishman. "Relation Extraction: Perspective from Convolutional Neural Networks."

• Kim, Yoon. "Convolutional neural networks for sentence classification." arXiv preprint arXiv:1408.5882 (2014).

• Zeng, D., Liu, K., Lai, S., Zhou, G. and Zhao, J., 2014, August. Relation classification via convolutional deep neural network. In Proceedings of COLING (pp. 2335-2344). Vancouver