Skip to content

Latest commit

 

History

History
25 lines (21 loc) · 1013 Bytes

File metadata and controls

25 lines (21 loc) · 1013 Bytes

Bi-lingual-location-reference-identification

Deep neural networks for location reference identification from bi-lingual disaster-related tweets

This repository contents a Python notebook for the location reference identification from bi-lingual English-Hindi tweets. Required packages: Python Keras Tensorflow Pandas Colne Fasttext: !git clone https://github.com/facebookresearch/fastText.git Sklearn Python library

#This repository contains, one Python notebook, weights for the trained models, and confusion matrices (five different weights and five different confusion matrices, as the model, is trained using 5-fold cross-validation).

If you are using this code, please cite:

@article{kumar2021location, title={Deep neural networks for location reference identification from bi-lingual disaster related tweets}, author={Kumar, Abhinav and Singh, Jyoti Prakash}, journal={IEEE Transactions on Computational Social Systems}, volume={volume}, pages={pages}, year={2022}, publisher={IEEE} }