Hey,
Awesome ideas dude.
First of all, I think लहान has forms [used google - search "लहानी in
marathi"] and its just that they are not normally used. Correct me if I am
wrong, its been 6 years since I have used Marathi.
Check this, its a shallow parser[intermediate output has morph analysis
too] for Marathi: http://ltrc.iiit.ac.in/analyzer/marathi/ . This is the
online interface one. They have a downloadable version too. But, that has
some bugs and it didn't work for me. Try it and tell me if it works for you.
The apertium-dixtools is not working for me. Can you send your tools or
version number?
- Nurendra Choudhary.
On Mon, Jan 26, 2015 at 12:57 AM, Vinit Ravishankar <
[email protected]> wrote:
> Hey Nurendra,
>
> I've got a large list of words, sorted by frequency, but, as you said,
> there's likely to be a lot of error with us manually conjugating/declining
> them.
>
> A couple of (probably terrible) suggestions - how about we go
> alphabetically, adding pardefs if the words don't match already existing
> pardefs? It's easier to keep track of pardefs than doing it more randomly;
> I noticed it was pretty helpful, when I was looking at the German monodix.
> Second, apertium-dixtools makes adding words pretty simple too - you enter
> a word that conjugates similar to the one you want to add, and it looks up
> the pardef for that word (if it's already in the dictionary), and adds the
> new word. I had added a batch mode to it, I'm not sure if it's up, though.
>
> Also, a minor mistake I noticed - the pardef for 'लहान'; I don't think the
> word has any other forms, just 'लहान'. I'll mail a few more words tomorrow
> :)
>
> Somebody else who's worked on monodices - could you give us a few
> pointers?
>
> Cheers,
> Vinit
>
> On Sun, Jan 25, 2015 at 2:55 PM, Nurendra Choudhary <
> [email protected]> wrote:
>
>> Hi,
>> Thanks again for checking out the errors. I am presently not using any
>> system for pardefs but just taking words and getting their morph analysis
>> manually. That is the reason, I am pretty slow currently and maybe not
>> accurate. I have tried some systems but none of them work properly.
>> I would appreciate it if you could suggest me some systems for Marathi.
>> Committed your errors.
>>
>> -Nurendra Choudhary.
>>
>>
>> ------------------------------------------------------------------------------
>> New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
>> GigeNET is offering a free month of service with a new server in Ashburn.
>> Choose from 2 high performing configs, both with 100TB of bandwidth.
>> Higher redundancy.Lower latency.Increased capacity.Completely compliant.
>> http://p.sf.net/sfu/gigenet
>> _______________________________________________
>> Apertium-stuff mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/apertium-stuff
>>
>>
>
>
> ------------------------------------------------------------------------------
> New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
> GigeNET is offering a free month of service with a new server in Ashburn.
> Choose from 2 high performing configs, both with 100TB of bandwidth.
> Higher redundancy.Lower latency.Increased capacity.Completely compliant.
> http://p.sf.net/sfu/gigenet
> _______________________________________________
> Apertium-stuff mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/apertium-stuff
>
>
------------------------------------------------------------------------------
New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
GigeNET is offering a free month of service with a new server in Ashburn.
Choose from 2 high performing configs, both with 100TB of bandwidth.
Higher redundancy.Lower latency.Increased capacity.Completely compliant.
http://p.sf.net/sfu/gigenet
_______________________________________________
Apertium-stuff mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/apertium-stuff
