[developers] color adjectives ARG1: *top*

Michael Wayne Goodman goodmami at uw.edu
Tue Mar 31 15:32:20 CEST 2015


Great, thanks for checking, Sanghoun!

On Tue, Mar 31, 2015 at 6:19 PM, Sanghoun Song <sanghoun at uw.edu> wrote:
> Hi developers,
>
> Using my versions (the trunk version of ERG, 0.9.19 of ACE), I got i13, not
> *top*.
>
> [ LTOP: h0
> INDEX: e2 [ e SF: prop TENSE: pres MOOD: indicative PROG: - PERF: - ]
> RELS: < [ _the_q_rel<0:3> LBL: h4 ARG0: x3 [ x PERS: 3 NUM: sg ] RSTR: h5
> BODY: h6 ]
>  [ "_color_n_1_rel"<4:9> LBL: h7 ARG0: x3 ]
>  [ _be_v_id_rel<10:12> LBL: h1 ARG0: e2 ARG1: x3 ARG2: x8 [ x PERS: 3 NUM:
> sg ] ]
>  [ udef_q_rel<13:18> LBL: h9 ARG0: x8 RSTR: h10 BODY: h11 ]
>  [ "_blue_a_1_rel"<13:18> LBL: h12 ARG0: x8 ARG1: i13 ] >
> HCONS: < h0 qeq h1 h5 qeq h7 h10 qeq h12 >
> ICONS: < > ]
>
> I tried to use the ERG 1212, and then got the same result as Michael. I
> think the latest version of ERG has fixed this problem.
>
> Sanghoun
>
>
> On Tue, Mar 31, 2015 at 5:57 PM, Michael Wayne Goodman <goodmami at uw.edu>
> wrote:
>>
>> Hi Woodley and Dan (and developers, because openness),
>>
>> I tried parsing a sentence like "The color is blue." with the ERG
>> (1212) and ACE (0.9.20), and I get an MRS like this for the second
>> result (which I think is the preferred parse):
>>
>> [ LTOP: h0
>> INDEX: e2 [ e SF: prop TENSE: pres MOOD: indicative PROG: - PERF: - ]
>> RELS: < [ _the_q_rel<0:3> LBL: h4 ARG0: x3 [ x PERS: 3 NUM: sg ] RSTR:
>> h5 BODY: h6 ]
>>  [ "_color_n_1_rel"<4:9> LBL: h7 ARG0: x3 ]
>>  [ _be_v_id_rel<10:12> LBL: h1 ARG0: e2 ARG1: x3 ARG2: x8 [ x PERS: 3
>> NUM: sg ] ]
>>  [ udef_q_rel<13:18> LBL: h9 ARG0: x8 RSTR: h10 BODY: h11 ]
>>  [ "_blue_a_1_rel"<13:18> LBL: h12 ARG0: x8 ARG1: *top* ] >
>> HCONS: < h0 qeq h1 h5 qeq h7 h10 qeq h12 > ]
>>
>> Notice the value of ARG1 on _blue_a_1_rel is *top*, which I don't
>> think is valid in an MRS.
>>
>> I'm not sure if this is an ACE or ERG bug. The 1214 release used on
>> erg.delph-in.net has an i-variable instead of *top*, which is what I
>> expect.
>>
>> I think I can mostly ignore this for my own purposes, but it's worth
>> noting here in case it is a bug.
>>
>> --
>> -Michael Wayne Goodman
>
>
>
>
> --
> ====================================
> Sanghoun Song
> Ph.D. in Computational Linguistics | http://corpus.mireene.com
> NTU Computational Linguistics Lab. | http://compling.hss.ntu.edu.sg
> ====================================



-- 
-Michael Wayne Goodman


More information about the developers mailing list