LT is using its own tagger. You can also try it here: Text Analysis - LanguageTool. It cannot easily be replaced, as all the rules now rely on its output.
I mean only change the part of tag word (Part-of-speech). Could we use stanford-posttagger ? and it’s output every word with only one token.
The Stanford tagger is GPL, thus we cannot use it. But the main issue is that we have 1000 rules that were developed with the existing tagger, they would all have to be checked and maybe rewritten for a new tagger.
ohh. Thanks anyway.
How to add this word(prosumer) in the dict of LT?
You can add new words for the English spell checker in this file:
languagetool-language-modules/en/src/main/resources/org/languagetool/resource/en/hunspell/spelling.txt
Thanks, I want to change the message in EnglishModel into Chinese, but
It appears garbled. How should I do?
What exactly do you mean by “EnglishModel”? Have you saved all modified files in UTF-8?
Thanks, I have solved this problem. Today, I test LT(below) and have a new question:
Input: I afraid of the dark.
Output:
Potential error at line 0, column 323: You miss the verb here. You should rather write: I’m afraid.
Suggested correction: [I’m afraid]
and,
Input: I’m afraid the dark.
Output:
Potential error at line 0, column 327: The adjective afraid is used with ‘of’: afraid of the dark.
Suggested correction: [afraid of the dark]
but,
Input: I afraid the dark.
Output:
Potential error at line 0, column 325: The adjective afraid is used with ‘of’: afraid of the dark.
Suggested correction: [afraid of the dark]
Why only one mistake be found?
When running the tests, for each example only one rule is active (the rule the example belongs to), so you cannot have two errors in one example.