Definition part of speech of the word?

0 like 0 dislike
3 views
What are algorithms to determine the part of speech of the word?


In addition to listing the endings in the head did not come.


/*
Group endings:
1. adjectives
2. communion
3. Glagol
4. noun
*/

$groups = array(
1 => array ('ie','ie','ie','OE','IMI','UMI','a','I','y','EU','am','im','im','am','ego','ego','EMU','EMU','a','YH','uu','uu','a','AA','OU','AU'),
2 => array ('IVS','yws','ous','eat','NN','VSH','yusch','u'),
3 => array ('ILA','ILA','EN','ate','white','ITE','ily','Ali','a','Wi','Il','Il','im','im','in','ILO','ILO','ene','at','Wat','wut','it','it','any','it','it','ISL','uu','u','l','na','ate','ITE','Li','I','l','em','n','lo','no','et','ut','NY','ti','ESL','no'),
4 => array ('a','EW','ow','EE','ie','e','EMI','EMI','EMI','ei','ie','I','ia','a','AI','I','I','em','am','am','am','am','am','a','u','Ah','IAH','IAH','y','y','IU','u','u','ia','ia','a'),);
by | 3 views

7 Answers

0 like 0 dislike
Look at the suffix/ending is not reliable, compare:
\r
rocks (b. p.) — General — knew

These words belong to verbs or to nouns?
\r
I think option two: or dictionary with word forms (eg. dictionary of Zaliznyak) and look for it (but homonyms and homophones are a problem: the glass (noun) — glass (CH., PR. V.), etc.), or to try to implement a simple parser, but then the problem is a loose word order in Russian.
\r
In any case, if you need accuracy, just to realize it will not work.
by
0 like 0 dislike
Here the link to the PDF about Mistem — company.yandex.ru/technology/mystem
by
0 like 0 dislike
The third option (hybrid :)):
1) look in the dictionary,
2) if the word is not found, trying to get the prefix from the beginning of the word and look again,
3) if there were a few options, do the parsing.
by
0 like 0 dislike
yeah... Russian is a noun or an adjective?
If you need a really reliable result — will have to think about the context.
by
0 like 0 dislike
doing similar, fasteners, unbearable, rye
by
0 like 0 dislike
a programming language is what? depending on the language and needs looking at phpmorphy, lemmatizer, pymorphy, mystem.
by
0 like 0 dislike
On aot.ru there is lemmatiser, which makes morphological analysis.
IMHO, the most appropriate and functional tool for this kind of task.
by
110,608 questions
257,186 answers
0 comments
28,647 users