Closed obtic-sorbonne closed 4 years ago
Thanks for reporting. Can you provide a suitable example I can use to reproduce the problem? Ideally, a .txt file and a word or phrase suggestion?
Please find attached a sample file (or here) from online news. You can annotate the word فيروس (virus). The system will suggest you:
There are 3 more un-annotated occurrences of فيروس in the text. Do you want to re-apply this annotation?
Actually, the 3 others occurences are agglutinated to prefixes: الفيروس الفيروس بالفيروس
Thanks for your help! ps. you can find the same example here: corona news.txt
Hi,
just a quick heads that I finally got a chance to look into this (and an apology for this taking so long...) I think I can publish a fix later today. Will let you know.
Cheers, R
The fix is now applied to production. Do let me know if the problem persists for you. Otherwise: happy annotating ;-)
Thank you very much, it works well !
When I use:
It works well for French but not for Arabic (RTL) because it annotates any string match, even inside words. Thanks for your help.