Closed MacMachin closed 7 years ago
@MacMachin These are not errors. The statistics below the POS guidelines are automatically computed. As you say, there are orthographic errors in the original texts, but we annotate the data with the intended POS. So if "à" is used as a verb, it is tagged VERB.
@mcdm What a silly answer ! These are not errors but there are errors ! Your parser is not able to detect errors and you are proud of that ? And you want to keep them ? If your data are not reliable they are not useful. You should better correct them.
@macmachin I think you are misunderstanding things here. The statistics are just counts, not results from a parser. We are doing corpus annotation, and we keep the text as is: if the text is incorrect, we don't modify it.
On Oct 6, 2016, at 3:15 AM, MacMachin notifications@github.com wrote:
@mcdm What a silly answer ! These are not errors but there are errors ! Your parser is not able to detect errors and you are proud of that ? And you want to keep them ? If your data are not reliable they are not useful. You should better correct them.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.
There are errors in the page ADP for French language due to orthographic errors in the initial texts:
AUX
or aVERB
, theVERB
is "a" without the accent ;NOUN
, in the example it should be written with quotes because the text speaks about the word "à" ;VERB
, theVERB
is "part" or "pars" ;ADJ
but not in the example given where it should be written "sûr" with a circumflex accent ;