- An N-gram statistical model to predict next word based on input words.
- The data sets for this project is given by Coursera-SwiftKey.
en_US(English) is used to for this project.- The data sets is cleaned and N-gram tokens are generated.
- The preprocessing step can be reproduced here.
2023-09-08
Next Word Prediction Application
N-gram tokens Visualisation
The UI of the web app.
The Application and its working
- The application uses ngram statistical modelling, meaning it is based on the frequency of the word usage the next word is predicted.
- A very small percentage of the data set is taken as the sample. So the accuracy is naturally affected.
- More the sample size, better the accuracy of prediction model.