shohrehhd / persian_pos_tagger Goto Github PK

"#persian_pos_tagger" "This is an implementation of hmm-based persian tagger" This code runs in 4 modes :

1-Training Mode: to run this mode you need Bijankhan-corpus ,you can find it at: http://dadegan.ir/catalog/bijankhan

2-Test Mode:

in this mode a 10-fold(90%-10%) cross validation will run on the Bijan-Khan corpus

3-Label Mode:

in this mode you can input a sentence and the program will print the pos-tags in rthe output.

4-Label mode:(file)

in this mode you input the adress of a file in which words are sepereated by line break or space or tab and the program will create a file with pos-tags besides each word. the outputfile is named: @inputputfilename+"_outfile"

Shohreh Haddadan