Vietnamese spell corrector and text classifier

Demo

You have to download the FULL model( including data) for the classifier in model and put it to our folder

Or the whole project

You also should take a look at our approach and implementation are presented in the Presentation.pptx

Windows

• Open your Command Prompt

>> cd <path to the folder which contains the app.py file>
>> pip install -r requirements.txt
>> set FLASK_ENV=development
>> flask run 

Then take the address and open this link on your browser

A/Spell corrector

• Data:
  • VNTC
  • Wiki article
• Prepocess text, build the neural network and evaluate the accuracy:
  • Training phase
• Test the model:
  • Put your text to the add_noise.py file and it will return the corresponding noisy text
  • You can also edit that text like the way you want
  • Then put the noisy text to spell corrector program (on your browser) and see the result
• Result

B/Text classifier

• Data:
  • VNTC
• Test the model:
  • Put the Vietnamese text to text classifier program (on your browser) and see the result
• Result

Many thanks to Lê Mạnh Hùng for helping us to finish this project