Using CAMeL Tools to normalise Egyptian Arabic symptom descriptions (veterinary) before mapping to English clinical terms

[vetegyai-ops]

Hi CAMeL Lab team, thanks for CAMeL Tools.

I'm building a veterinary assistant for Egyptian farmers and pet owners. They write symptoms in Egyptian colloquial Arabic, for example "الفراخ بتنهج وفاتحة بوقها" (chickens are panting with open beaks), "الجاموسة مانعة الأكل" (the buffalo refuses feed) and "عنده اسهال بقاله يومين" (diarrhoea for two days). The medical evidence is in English. Word-level dictionaries (AGROVOC and a ~300-term field glossary) did not help much: they miss inflected dialect forms such as بتنهج → نهج and بقاله, and they add generic noise.

Questions:

1. For this use case, would you recommend the EGY morphological analyzer/database (CALIMA-EGY) for lemmatising dialect verbs and nouns before lexicon lookup? Is there an example pipeline: orthographic normalisation → dialect ID → EGY analysis → lemma?
2. How reliable is dialect identification on short (5–20 word) farmer messages?
3. Do you know of any Arabic or Egyptian medical or veterinary symptom lexicon, or annotated corpus, that works with CAMeL Tools?
4. Any known issues running on Linux servers (we noticed DID is not available on Windows)?

Thanks!