Good day. I am a student from the Philippines, part of my research is to preserve the language of the tribe in our province by creating a grammar checker of their language. I am new to NLP. I have seen the grammar checker you guys created including the tagalog language. I am here for my research to get some advice on how to start building a grammar checker for another language for the goal of preserving culture of this tribal language.
I am currently studying the NLTK book since python is easier to understand, doing java is fine for me also as i have seen the code of languagetool. But my main problem here is how do i create the parts of speech and its tree. Learning this stuff would be great and if its fine I would like to join the community to learn more.
To start with LT, you do NOT necessarily need a postag dictionary. You could simply start with a speller dictionary and rules in XML. That is how I started over 10 years ago for Dutch. There is part on ‘starting a new language’ on the wiki.
But I would like to say that LT is not a teaching tool; it is a (very flexible and good) writing aid.
I assume the grammar of the language has already been documented? And there is an extensive words list for the language as well?
Having electronic real-life texts in the language also helps for testing.
I will start documenting the grammar tomorrow and also the words. What is this XML. Is NLTK a good start? and is it possible to insert foreign language in this case ibaloi language to nltk together with its grammar?