Spelling Corrector

PROJECT TITLE : SPELLING CORRECTOR

AIM

To correct spellings of words

DESCRIPTION

This is a problem where we need to correct the spellings of different misspelled words. We use Sequence to Sequence Encoder-Decoder Model Model.

LINK TO WEBAPP:

https://share.streamlit.io/shreya024/spellcorrector/main/app.py

GLANCE AT THE WEBAPP

spelling_corrector.mp4

MODELS USED

Sequence to Sequence Encoder-Decoder Model - 99.12 % accuracy

LINK TO MODEL FILES:

https://drive.google.com/drive/folders/1mpvIWLqvsakD7FEIDIoU7-APbo1f077b?usp=sharing

DEPLOYMENT

StreamLit-Share = Streamlit turns data scripts into shareable web apps in minutes. All in Python. No front‑end experience required. Streamlit’s open-source app framework is a breeze to get started with. It is very easy to use. One just have to connect with the github repo and start deploying it on streamlit share and you are good to go.

LIBRARIES NEEDED

1. Numpy
2. Tensorflow
3. Keras
4. Joblib
5. re
6. Unicode
7. StreamLit

CONCLUSION

We can conclude that our Web App corrects misspelled words with the help of Sequence to Sequence Encoder-Decoder Model.

REFERENCES

Some utilities functions used in this project are inspired by this medium post and from this Deep Spell code, with several distinct changes, additions and customization in this implementation, that result in a better performance.

CONTRIBUTED BY

Shreya Ghosh