Skip to content

8sukanya8/SpellingCorrections

Repository files navigation

SpellingCorrections

Aim: To detect incorrect words and use one or more models to find suitable corrections for incorrect words

main

Tasks:
[1] Coordinates the creation of the word dictionary
[2] Creates training and test sets
[3] Executes training and builds up all the models
[4] Compares performance of all models on the test set

Noisy Channel Model

Contains all the functions for training according to the Kernighan, Church, Gale probabilistic Noisy Channel

Unsupervised Rules Model

Contains all the functions for training according to Unsupervised rules in Soo and Frieder

N gram characters Model

Contains all the functions for training according to the Ngram model

UnsupervisedNgram Model

(Combination of Unsupervised Rules and N gram characters)

How to run?

python3 main.py

About

detect incorrect words and use one or more models to find suitable corrections for incorrect words

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages