Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

new test sentences #155

Merged
merged 12 commits into from
Aug 28, 2014
Merged

new test sentences #155

merged 12 commits into from
Aug 28, 2014

Conversation

kkasunperera
Copy link
Contributor

@linas @ruiting Please review the newly added test sentences.

@@ -948,6 +948,45 @@ public boolean test_equatives()
"as(run, cheetah)\n"+
"_advmod(fast, exactly)\n"+
"than(snail, cheetah)\n");

//one entity one feature, through time
rc &= test_sentence ("The coffee tastes as same as it did last year.",
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please change this to "tastes the same" not "tastes as same"

@linas
Copy link
Member

linas commented Aug 12, 2014

I don't want to merge this right now; there seem to be far too many errors in this ... can you please fix?

"_predadj(he, smart)\n"+
"_subj(expect, I)\n"+
"_obj(expect, him)\n"+
"as(smart, expect)\n"+
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

same question as above should it be as(smart, I)?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

as(smart, I) doesn't seem right either.. I'm confused about your decision on representing the "as" comparatives. I mean, for the above sentences, sometimes you used "than", sometimes you used "as", what's your criteria on that?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@ruiting Well my intention was to use "as" as a preposition only, and "than" in comparatives. Please pointout any relationship doesn't adhere to this.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In src/java_test/relex/test/TestRelEx.java:

  •       "_subj(run, Mike)\n"+
    
  •       "as(fast, he)\n"+
    
  •       "_advmod(run, fast)\n"+
    
  •       "_amod(year, last)\n"+
    
  •       "than(Mike, he)\n");
    
  •   rc &= test_sentence ("The kick was as soft as the first.",
    
  •       "_predadj(kick, soft)\n"+
    
  •       "as(soft, first)\n"+
    
  •       "than(kick, first)\n");
    
  •   rc &= test_sentence ("He is as smart as I ever expected him to be.",
    
  •       "_predadj(he, smart)\n"+
    
  •       "_subj(expect, I)\n"+
    
  •       "_obj(expect, him)\n"+
    
  •       "as(smart, expect)\n"+
    

There seem to be multiple issues here.

First of all, if the word "than" does not actually appear in the sentence,
then you can't use the relationship "than", you have to use "_than" with
the underscore.

Next, I'm pretty sure that the above should not have _than it them anyway.
I don't see why it should.

Next, I'm thinking that the "as" relationship should connect the main verb
of the dominant clause to the main verb of the dependent clause. It
should connect to the main noun of the dependent clause only if the verb is
unvoiced,

so above;
as(taste, do)
as(run, do)
as(soft, first)
as(smart, be)

right?

what does the stanford parser do for these cases?

//one entity one feature, through time
rc &= test_sentence ("The coffee tastes the same as it did last year.",
"_subj(taste, coffee)\n"+
"_obj(taste, same)\n"+
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have added _obj(taste, same) here. "taste" and "same" are connected through a "O" connector and "same" is acting as a prepositional object. Hope it's correct?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

_obj(taste, same) doesn't seem correct to me.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@ruiting I was thinking that DD connector(which is between "the" and "same") can be used to connect determiners to adjectives when they
are being used as self-contained noun-phrases. In this case "the same" as a object. another example is "It applies to the poor" I got this idea from here http://www.abisource.com/projects/link-grammar/dict/section-DD.html (see last two paragraphs)

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

_obj(taste, same) doesn't seem correct to me.

Right. This is a bug in link-grammar. Kasun, please change the test
sentence to "the coffee looks the same as it did last year". Somehow,
tastes/smells does not parse correctly. Also, please open a bug for me,
reminding me to fix this sentence (and that the fix involves taste->look)
-- I don't have github access this week.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@linas I just opened a new issue for the copula bugs. #160

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I just pushed a fix for this in LG, so #160 is now closed.

@kkasunperera
Copy link
Contributor Author

@linas I think it's good to merge now? unless @ruiting disagree on anything?

linas added a commit that referenced this pull request Aug 28, 2014
@linas linas merged commit ef8a0ef into opencog:master Aug 28, 2014
@linas
Copy link
Member

linas commented Aug 28, 2014

thanks Kasun

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants