Greetings all,

The ESPDIC Esperanto-English dictionary project has just been updated with some additons and corrections, bringing the total entry count to over 31,000.

http://www.denisowski.org/Esperanto/Esperanto.html

I've written a small program to help expand the dictionary by parsing through Esperanto text files and creating a list of words not found in ESPDIC. I ran it against the Project Gutenberg etext of Robinsono Kruso and it produced a list of ~600 words not in ESPDIC. After adding these entries and re-running the program, it produced at 98% match rate : almost every word in the etext could be found in ESPDIC (including such useful terms as kaprograso : goat fat).

As part of this ongoing project I plan to work through Gutenberg etexts as well as other Esperanto text files I come across. Hopefully this will provide a rich source of entries in addition to manual updates and contributions from others.

Kore,

Paul

P.S. One non-Gutenberg text I plan on using is La Sankta Biblio (1926, Brita kaj Alilanda Biblia Societo). I’ve also formatted a version of this text for the Kindle and other ebook readers : http://www.denisowski.org/Esperanto/Biblio/Bibli...

That is very impressive brainpower.

Kudos!

Hej pdenisowski,

do you have any idea how to create an rimvortaro which works similar like an spellchecker? I.e. someones writes an sentence and that ryhme spellchecker could give suggestions which words rhymes to words of that sentence.

There excist an online rimvortaro which is an extract of the well-knownreta-vortaro.de . That rimvortaro is called Rata Vortaro. There's also an pdf version of that. It can be found here. Look for "Rata_Vortaro_18_aprilo_2005.pdf". I will try to convert that into an MSWord and OpenOffice document later. It will be available at the same location.

I'm not sure if that works, but first I had an XML solution in mind. So the rhyme suggestions will pop up like seen here at this screencast for the XML vocabulary dita. Until now I failed to find an XML rhyme vocabulary suggestion example. So that the ryhme suggestions are inside an special XML vocablary file. That would be great, because there are many free XML editors out for usage with that customized XML vocabulary. I.e. the XMLmindEditor is free for personal use, too. But there are much more other free XML editors out. I.e. Komodo Edit etc.

Anyway, another possiblity could be to "mis-use" an normal spellchecker. That let me trip over the dictionary builder of XMLmind which compiles special Java spellchecker files. I tried to struggle through the topic, but it's not that easy I believed. Maybe I can figure out later.

Generaly, it would be fine to have modern rimvortaro tool available. There are probably a lot of e-o musicans out who would high appreciate such tool to write high-quality lyrics for their musics.

ĝp,

qwertz:Hej pdenisowski,

do you have any idea how to create an rimvortaro which works similar like an spellchecker? I.e. someones writes an sentence and that ryhme spellchecker could give suggestions which words rhymes to words of that sentence.

There excist an online rimvortaro which is an extract of the well-knownreta-vortaro.de . That rimvortaro is called Rata Vortaro. There's also an pdf version of that. It can be found here.

Generaly, it would be fine to have modern rimvortaro tool available. There are probably a lot of e-o musicans out who would high appreciate such tool to write high-quality lyrics for their musics.

ĝp,

It would be pretty simple to write a program to do this, but I don't really have the bandwidth or motivation to do it myself. The tool I wrote (in Perl) just makes a list of words found in the source text (removing numbers, punctuation, extra whitespace, etc.), then converts the words to a "base form" -- i.e. removes -j, -n for adjectives/nouns/adverbs (virojn becomes viro, hejmen becomes hejme), and converts most verb and particle endings to the infinitive (-as/-is/-os/-us/etc. become -i). There are a few cases where the tool misconverts words : kaj becomes "k", dekdu becomes dekdi, etc. but the misconvert rate is so low I haven't put in any special checking for words that don't follow standard Esperanto morphology.

If there's enough interest, I'll clean up (and comment) my Perl script and post it to the ESPDIC page as well.

Thanks,

Paul

Thanks Paul for that information.

Maybe its also interesting for you: Vortaroj de lernu! en via telefono.

ĝp,