phquang / vn_punctuation

Vietnamese Punctuation Prediction using Conditional Random Fields

---

Requirement

• Dowwnload and follow the instruction to install CRF++ [github] .

Data & Template

• The sample data sets and templates are provided in the news\ and subtitle folders

Experiment

1. Run ./bash_train to train 3 models with the provided templates and run prediction on the test data.
2. To calculate the F-score, run F-score.py result1