Named-Entity-Recognition-with-Bidirectional-LSTM-CNNs

A keras implementation of Bidirectional-LSTM_CNNs for Named-Entity-Recoganition. The original paper can be found at https://arxiv.org/abs/1511.08308

The implementation differs from the original paper in the following ways :

lexicons are not considered
Bucketing is used to speed up the training
nadam optimizer used instead of SGD

Result

The model produces a test F1_score of 90.9 % with ~70 epochs. The results produced in the paper for the given architecture is 91.14 Architecture(BILSTM-CNN with emb + caps)

Dataset

conll-2003

Network Model in paper

Network Model Constructed Using Keras

To run the script

   python3 nn.py

Requirements

1) numpy 
2) Keras
3) Tensorflow

kai04/Named-Entity-Recognition-with-Bidirectional-LSTM-CNNs