Deep Learning approaches for Named Entity Recognition
Submitted by Vijay Ramakrishnan (@vijay120) on Thursday, 13 July 2017
Section: Crisp talk Technical level: Intermediate
Using a Bi-Directional LSTM network we were able to achieve state of the art accuracies on Named Entity Data on WNUT 2016 Twitter Noisy data.
- Introduction on Named Entity Recognition and why it is important for voice assistants
- Introduction on prior approaches taken for NER
- Describe the LSTM model and how it’s better than existing approaches
- Demo the model on Tensorflow
Vijay Ramakrishnan is a machine learning engineer at Cisco, after his company, Mindmeld, was acquired by Cisco in June 2017. He is interested in deep learning approaches to NLP and self driving applications.