|
Welcome to the UNL Forum
|
|
|
|
View previous topic :: View next topic |
Author |
Message |
deva3105
Joined: 11 May 2011 Posts: 4
|
Posted: Wed May 25, 2011 7:12 am Post subject: Dictionary inconsistency? |
|
|
Quote: |
[write]{161413} "201744611" (LEX=V, SEM=CRE, POS=VER, LST=WRD, TRA=TST, FRA=Y32, PAR=M1, FLX(PAS:="write">"wrote";PTP:="write">"written";GER:=1>"ing";3PS&PRS:=0>"s";)) <eng>; |
The above dictionary entry says the verb is transitive, but gives the framing as Y32. According to the English Reference grammar, Y32 is defined as follows.
Quote: |
Y32 = VS(NP); (Intransitive verb: walk away, make peace, kill oneself) |
I was under the impression transitive verbs have a framing like that of Y38, i-e with a VS and a VC.
Quote: |
Y38 = VS(NP)VC(NP); (Direct transitive verb: accept, accomplish, acknowledge) |
I observed this with many other entries as well.
Is this an error in the dictionary or some error in my understanding?
Thanks,
Deva |
|
Back to top
|
|
|
martins Site Admin
Joined: 16 Dec 2009 Posts: 1481 Location: Geneva, Switzerland
|
Posted: Wed May 25, 2011 2:08 pm Post subject: RES: Dictionary inconsistency? |
|
|
Dear Deva,
Thank you for the observation. You’re right. There are indeed many inconsistencies yet in the English dictionary, which has been inherited from the WordNet and is still under revision. Actually, TST has been used as a sort of default value in case of homonymy and polysemy, because it subsumes several other values (TSTD, TSTI, etc). Notice, for instance, that, in the English dictionary, there are several entries “write” (see below). Most of them are FRA=Y38. Others are classified as FRA=Y56, FRA=Y39, FRA=Y32, etc., depending on the case. Until we can further analyze them, we have decided to keep them as transitive, which is the non-marked value for transitivity in the UNLarium framework.
Best regards,
---------------------------------------------
Ronaldo MARTINS
Language Resources Manager
UNDL Foundation
48, route de Chancy
CH-1213 – Geneva - Switzerland
+41 22 879 8090
http://www.undlfoundation.org
---------------------------------------------
155475
write [Edit] [Problem] [Inflections] [Lexical relations] [Translations] [Author]
communicate by letter ( = write )
FRA=Y56; LST=WRD; PAR=M1; POS=VER; TRA=TST;
FLX(PAS:="write">"wrote";PTP:="write">"written";GER:=1>"ing";3PS&PRS:=0>"s";)
155504
write [Edit] [Problem] [Inflections] [Lexical relations] [Translations] [Author]
= drop a line
communicate (with) in writing ( = write, drop a line )
FRA=Y39; LST=WRD; PAR=M1; POS=VER; TRA=TST;
FLX(PAS:="write">"wrote";PTP:="write">"written";GER:=1>"ing";3PS&PRS:=0>"s";)
155880
write [Edit] [Problem] [Inflections] [Lexical relations] [Translations] [Author]
communicate or express by writing ( = write )
FRA=Y38; LST=WRD; PAR=M1; POS=VER; TRA=TSTD;
FLX(PAS:="write">"wrote";PTP:="write">"written";GER:=1>"ing";3PS&PRS:=0>"s";)
155969
write [Edit] [Problem] [Inflections] [Lexical relations] [Translations] [Author]
= save
record data on a computer ( = write, save )
FRA=Y38; LST=WRD; PAR=M1; POS=VER; TRA=TST;
FLX(PAS:="write">"wrote";PTP:="write">"written";GER:=1>"ing";3PS&PRS:=0>"s";)
159702
write [Edit] [Problem] [Inflections] [Lexical relations] [Translations] [Author]
= compose, indite, pen
produce a literary work ( = write, compose, pen, indite )
FRA=Y38; LST=WRD; PAR=M1; POS=VER; TRA=TST;
FLX(PAS:="write">"wrote";PTP:="write">"written";GER:=1>"ing";3PS&PRS:=0>"s";)
159981
write [Edit] [Problem] [Inflections] [Lexical relations] [Translations] [Author]
= spell
write or name the letters that comprise the conventionally accepted form of (a word or part of a word) ( = spell, write )
FRA=Y38; LST=WRD; PAR=M1; POS=VER; TRA=TST;
FLX(PAS:="write">"wrote";PTP:="write">"written";GER:=1>"ing";3PS&PRS:=0>"s";)
161338
write [Edit] [Problem] [Inflections] [Lexical relations] [Translations] [Author]
= publish
have (one's written work) issued for publication ( = publish, write )
FRA=Y32; LST=WRD; PAR=M1; POS=VER; TRA=TST;
FLX(PAS:="write">"wrote";PTP:="write">"written";GER:=1>"ing";3PS&PRS:=0>"s";)
161831
write [Edit] [Problem] [Inflections] [Lexical relations] [Translations] [Author]
= compose
write music ( = compose, write )
FRA=Y32; LST=WRD; PAR=M1; POS=VER; TRA=TST;
FLX(PAS:="write">"wrote";PTP:="write">"written";GER:=1>"ing";3PS&PRS:=0>"s";)
162050
write [Edit] [Problem] [Inflections] [Lexical relations] [Translations] [Author]
mark or trace on a surface ( = write )
FRA=Y32; LST=WRD; PAR=M1; POS=VER; TRA=TST;
FLX(PAS:="write">"wrote";PTP:="write">"written";GER:=1>"ing";3PS&PRS:=0>"s";)
162061
write [Edit] [Problem] [Inflections] [Lexical relations] [Translations] [Author]
create code, write a computer program ( = write )
FRA=Y38; LST=WRD; PAR=M1; POS=VER; TRA=TST;
FLX(PAS:="write">"wrote";PTP:="write">"written";GER:=1>"ing";3PS&PRS:=0>"s";)
|
|
Back to top
|
|
|
deva3105
Joined: 11 May 2011 Posts: 4
|
Posted: Thu May 26, 2011 6:24 am Post subject: |
|
|
Dear Ronaldo,
Thanks for the clarification and the quick response. I also noticed the following in the corpus of "Le Petit Prince" provided on the UNL website. Here, the UW 200588888 is, according to the English dictionary, of framing Y32(No VC). But sentence uses 200588888 (English word "understand") as a transitive verb. Is this an error in the corpus? (I found a number of other such cases as well).
Quote: |
[S:0007]
{org:fr}
cette grande personne peut tout comprendre, même les livres pour enfants.
{/org}
{unl}
aoj(200588888.@ability.@entry, 109605289.@proximal)
cnt(00.@every.@thing, 106410904.@generic.@even)
pur(106410904.@generic.@even, 109917593.@generic)
obj(200588888.@ability.@entry, 00.@every.@thing)
{/unl}
[/S] |
|
|
Back to top
|
|
|
martins Site Admin
Joined: 16 Dec 2009 Posts: 1481 Location: Geneva, Switzerland
|
Posted: Thu May 26, 2011 3:09 pm Post subject: RES: Re: Dictionary inconsistency? |
|
|
Dear Deva,
Given that the definition for the UW 200588888 is “know and comprehend the nature or meaning of”, which is exemplified by constructions such “She did not understand her husband | I understand what she means”,
I would say that the problem is once again related to the English dictionary, rather than to the corpus.
Best,
---------------------------------------------
Ronaldo MARTINS
Language Resources Manager
UNDL Foundation
48, route de Chancy
CH-1213 – Geneva - Switzerland
+41 22 879 8090
http://www.undlfoundation.org
---------------------------------------------
|
|
Back to top
|
|
|
deva3105
Joined: 11 May 2011 Posts: 4
|
Posted: Fri Jun 17, 2011 10:42 am Post subject: |
|
|
Dear Ronaldo,
The multi-word expressions in the dictionary seem to be underspecified. There is no information as to what lexical category the second(and third) word belongs to. Though it seems obvious to humans, it is not machine tractable. Is there any means of obtaining this information outside the dictionary?
Quote: |
[[be] [adrift]]{164096} "201902783" (LEX=V, SEM=MOT, POS=VER, LST=MTW, TRA=ITST, FRA=Y31, #01(POS=VER, MOR=STE, PAR=M1, FLX(1PS&PRS:="am";2PS&PRS:="are";3PS&PRS:="is";1PP&PRS:="are";2PP&PRS:="are";3PP&PRS:="are";1PS&PAS:="was";2PS&PAS:="were";3PS&PAS:="was";1PP&PAS:="were";2PP&PAS:="were";3PP&PAS:="were";PTP:="been";)), #02(VC)) <eng>;
|
Also, what exactly does VH mean?
Quote: |
[[abide] [by]]{168907} "202542280" (LEX=V, SEM=SOV, POS=VER, LST=MTW, TRA=ITST, FRA=Y32, #01(POS=VER, MOR=STE, PAR=M17), #02(VH)) <eng>; |
Deva
Invention Labs
India |
|
Back to top
|
|
|
martins Site Admin
Joined: 16 Dec 2009 Posts: 1481 Location: Geneva, Switzerland
|
Posted: Mon Jun 20, 2011 12:17 pm Post subject: RES: Re: Dictionary inconsistency? |
|
|
Dear Deva,
Multiword expressions are represented in the dictionary as complex entries such as the ones referred to below. Notice that the dictionary brings, in the feature list, the features assigned to the complex entry and to any of its components, in the following format:
[[ENTRY1][ENTRY2][ENTRY3]…MAIN ENTRY] = (MAIN ENTRY FEATURES, #01(FEATURES OF THE ENTRY 1), #02(FEATURES OF THE ENTRY 2), #03(FEATURES OF T HE ENTRY 3, … )
For instance:
[[be] [adrift]]{164096} "201902783" (LEX=V, SEM=MOT, POS=VER, LST=MTW, TRA=ITST, FRA=Y31, #01(POS=VER, MOR=STE, PAR=M1, FLX(1PS&PRS:="am";2PS&PRS:="are";3PS&PRS:="is";1PP&PRS:="are";2PP&PRS:="are";3PP&PRS:="are";1PS&PAS:="was";2PS&PAS:="were";3PS&PAS:="was";1PP&PAS:="were";2PP&PAS:="were";3PP&PAS:="were";PTP:="been" ), #02(VC)) ;
The feature list:
LEX=V, SEM=MOT, POS=VER, LST=MTW, TRA=ITST, FRA=Y31
Corresponds to the main entry (be adrift), whereas
#01(POS=VER, MOR=STE, PAR=M1, FLX(1PS&PRS:="am";2PS&PRS:="are";3PS&PRS:="is";1PP&PRS:="are";2PP&PRS:="are";3PP&PRS:="are";1PS&PAS:="was";2PS&PAS:="were";3PS&PAS:="was";1PP&PAS:="were";2PP&PAS:="were";3PP&PAS:="were";PTP:="been" )
Corresponds to the first component (be), and
#02(VC)
Corresponds to the second component (adrift).
Most of these entries have not been revised yet, and it’s normal that you couldn’t find there all the necessary information yet, especially for complements. In “be adrift”, for instance, you have the information that “be adrift” is a verb, and that “be” is a verb, but there is no information available about the part of speech of “adrift”. Actually, this information was not available in the source database from which we have imported the English data, and that’s why it’s not there. In any case, we have tried, at least for the time being, to represent its syntactic role: we know that “adrift”, for instance, is the complement (predicative) of the verb to be because of “VC” (verb complement); in “abide by”, we know that “by” is still a part of the verb (i.e., it’s a phrasal verb) because of “VH” (verb head). The full list of the syntactic roles may be found at:
http://www.unlweb.net/wiki/index.php/Syntactic_roles.
Best regards,
---------------------------------------------
Ronaldo MARTINS
Language Resources Manager
UNDL Foundation
48, route de Chancy
CH-1213 – Geneva - Switzerland
+41 22 879 8090
http://www.undlfoundation.org
---------------------------------------------
|
|
Back to top
|
|
|
vengity Guest
|
|
Back to top
|
|
|
vengity Guest
|
|
Back to top
|
|
|
vengity Guest
|
|
Back to top
|
|
|
vengity Guest
|
|
Back to top
|
|
|
vengity Guest
|
|
Back to top
|
|
|
vengity Guest
|
|
Back to top
|
|
|
vengity Guest
|
|
Back to top
|
|
|
vengity Guest
|
|
Back to top
|
|
|
vengity Guest
|
|
Back to top
|
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You cannot download files in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|