This software aims to train an Hidden Markov Model that is able to correct the spelling error in small sentences, as for example tweets.
Specifically, it has been shared through this repository:
- the training procedure used to estimate the three matrices that characterize an HHM; (available here)
- a Graphical User Interface through which the best HMM trained can be used to correct the tweets; (available here)
- all the traning and testing sets used.(available here)
For any further information about the training procedure and the GUI check out the Report.pdf
(N.B. this code has been tested only on Ubuntu 18.04 LTS)