lexfeatures

A wrapper for encoding Lexical features. Simple handcrafted features helps your sequence to sequence model better tagging and chuncking accuracy even in noisy setting. Extremly helpful in case of social media datasets like twitter, reddit or snapchat

Run

#pip install lexfeatures-SaichethanReddy

from lexfeatures import *
r = encoder("Saichethan M. reddy 170101025")
print(r.encode())

Citation

@article{suman2021pay,
  title={Why pay more? A simple and efficient named entity recognition system for tweets},
  author={Suman, Chanchal and Reddy, Saichethan Miriyala and Saha, Sriparna and Bhattacharyya, Pushpak},
  journal={Expert Systems with Applications},
  volume={167},
  pages={114101},
  year={2021},
  publisher={Elsevier}
}