This python script uses Hidden Markov Models and the Viterbi algorithm to perform part-of-speech (POS) tagging on a given file. It can be invoked using the following command
python3 tagger.py -d <training files> -t <test file> -o <output file>
<training files> are plain text files that have already been tagged. Such files consist of many lines, each of which is formatted as word : tag.
<test file> is a plain text file that you wish to tag. Each line of this file must contain exactly one word/punctuation character (whitespace should not be included). For example, the text "Hello, world?" would be formatted as
"
Hello
,
world
?
"
The output format is exactly the same as the specification for the <training files>.