rootfinder
Languages and libraries
arabic morphology
SHOW HN2018
How HN took it
3 positive, 7 neutral or asking, 2 mixed, 4 critical, out of 16 comments about the project. Thread score -0.05.
Idea: 2 for, 2 againstDesign: 0 for, 1 againstPerformance: 0 for, 1 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Interesting, well-executed work on Arabic morphology.
From 3 comments
Can you please explain what the inputs and outputs are supposed to be? I'm familiar with the Arabic alphabet but I don't know the words
ZainRiz on HNThis is exciting to see. I am a Semitic philologist (Ph.D.) now breaking into the IT industry, and this sort of work is on my radar, though mostly with Hebrew and Aramaic.…
ilimilku on HNCongrats for such nice non-English project! It is always lovely to see computer linguistics hitting the front page :-)
caio1982 on HN
Transliteration added for readers without Arabic script.
From 2 comments
I added some basic transliteration in the output formatting stage so people who don't know Arabic script can understand the results. Accepting romanized input would require a…
tboyd47 on HNNo problem. By the way, I added some quick and dirty romanization in the output stage if you're interested. It in no way represents how the words sound, but it does make it…
tboyd47 on HN
Objections
Rules-based parsing may work as well or better than a neural network.
From 3 comments
I like where this is going. But I'm still a bit skeptical at the use of neural networks in this use case How big is the problem space? There's a limited set of roots and…
anonu on HNThis is an interesting project, although as others have mentioned the space of arabic words is (reasonably) bounded and an explicit parsing approach or something that uses known…
ztravis on HNI took some Arabic at university. It's a fascinating language. My impression was that the morphology is quite regular, I wonder how complicated an old fashioned, hand-written…
rvense on HN
Output for some words is ambiguous, e.g. jumla yields a root that also spells a common word.
From 1 comment
> Correct: ['جملة', 'جمل'] جملة means a "sentence" جمل can mean a Camel (pronounced "jamel" you can guess the origin of the english word) جمل can also mean "sentences"…
csomar on HN
Requests
Support for Latin-script input.
From 1 comment
Great work! Are there any plans for latin character support? I’m Turkish and I’d love to be able to play with the words we’ve borrowed from Arabic but unfortunately I don’t know…
ulucs on HN
Training data from treebanks and shared tasks.
From 2 comments
Nice work! If I'm not mistaken, the root requires morphological disambiguation, which may change depending on the context/phrase in which the word is observed. This is an active…
habeanf on HNThis is an interesting project, although as others have mentioned the space of arabic words is (reasonably) bounded and an explicit parsing approach or something that uses known…
ztravis on HN
Questions
What are the inputs and outputs?
From 2 comments
Can you please explain what the inputs and outputs are supposed to be? I'm familiar with the Arabic alphabet but I don't know the words
ZainRiz on HNBut what are the inputs? Are they words? And what are the outputs? I know nothing about arabic, so to my eyes certain wiggles - that aren't a 1:1 match - are "correctly" matched,…
pavel_lishin on HN
Alternatives named
Arabic reference site arabicreference.com and Bayyinah sarf generator.
From 2 comments
This is an interesting project, although as others have mentioned the space of arabic words is (reasonably) bounded and an explicit parsing approach or something that uses known…
ztravis on HNThis is really cool. Interestingly, Bayyinah (www.bayyinah.com) Institute has a site [1] for the opposite problem where you can generate the 10 morphological families from root…
nafizh on HN
Universal Dependencies treebanks and SIGMORPHON shared task data.
From 1 comment
Nice work! If I'm not mistaken, the root requires morphological disambiguation, which may change depending on the context/phrase in which the word is observed. This is an active…
habeanf on HN