uima-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Klügl <peter.klu...@averbis.com>
Subject Re: Using RUTA
Date Wed, 22 Mar 2017 16:34:15 GMT
Hi,


Am 15.03.2017 um 10:24 schrieb José Vicente Moyano Murillo:
> Great explanation Peter, we see all your comments.
> Maybe we are doing something wrong to accomplish our pourpouse.
>
> But regarding labels. Can we discuss this example?
>
> Document{ -> ADD(list, w1), ADD(list, w2), ADD(list, w3), ADD(list, w4),
> CREATE(Detection, "anchors" = list)} <- {
>         w1:CW{w1.ct=="A" }
>         %
>         w2:CW{w2.ct=="B" }
>         %
>         w3:CW{w3.ct=="B"  }
>         %
>         w4:CW{w4.ct=="A" };
>        };
>
> In our mind conjuntion rules are aplyed using something like this pattern:
>
> FOR (PERMUTATION(w1, w2, w3, w4))
>       IF(CONJUNCTION) -> THEN
>
>
> In other words we spect some kind of "interruption" to execute the THEN
> statement

There is not explicit interruption. The execution is much more
"brute-force" using the usual ruta rule behavior.

I just summarize my interpretation a bit although you probably know all
of that already:

There are two special language constructs: an inlined rule as condition
and a conjunct rule element. For validating the condition of the main
rule element (Document), the inlined rules need to be validated, meaning
one of the inlined rules sto match. So, Ruta starts by trying to apply
the first rule, which is the only rule in this example. The rule
consists of some conjunct rule elements. Therefore, the rule element
starts to apply each conjunct rule element separately and all rule
element matches are remembered/stored. If at least one rule element
match for each of the conjunct rule elements matched successfully, then
the actions of all successful rule element matches are executed. If the
inlined rule was able to match, then the actual rule element matched
successfully, and the rule matched successfully, and the actions of the
actual rule are executed.


> But it seems that right now, all the permutations are inspected and then
> the necessary annotations are created and, obviously, w1, w2, w3, w4 have
> the values of the last iteration.

Yes, since the matching conditions of the conjuncted rule elements are
equal, the successful rule element matches should build all
permutations. The variables of the labels (w1,w2...) have the value of
the last match of the rule element even if it was not successful, the X.

> What do you think?
>
>

I am still wondering about what should/can be done to improve this
situation/use case. If I override the value of  the label variables with
null for an unsuccessful match, it would not solve the problem in this
current example. Resetting the value to the last valid one makes only
sense for the conjunct rule, but will be very problematic in all other
cases.

What I previously did in such cases, was to introduce a new language
construct that supports the current need, e.g., the conjunct rules.
Right now, the expressiveness of ruta is high enough that this should
not happen too often. On the other hand, maybe stepping back a bit and
looking at the problem to be solved from a different perspective is a
better approach. Maybe this should not be solved with labels and
conjunct rules, but maybe rather with something else like
ONLYONCE/ONLYFIRST, a boolean variable, or a new inlined rule....

Best,

Peter

>
>
>
>
>
>
> 2017-03-13 18:05 GMT+01:00 Peter Klügl <peter.kluegl@averbis.com>:
>
>> Hi,
>>
>>
>> Am 13.03.2017 um 17:41 schrieb José Vicente Moyano Murillo:
>>> From our point of view we would expect the same result using # or %
>> Yes, I thought the same, but after investigating the problem I changed
>> my mind a bit.
>> Even if we ignore the label expression overriding the values with 'X's,
>> there is still a difference. The wildcard # introduces a sequential
>> constraint, and the conjunction % does not. The former implies a match
>> only for the next occureence of a valid annotation. And the latter? I
>> would initially say that the match should not be restricted to the first
>> one, meaning the first rule element for example will match twice as does
>> the fourth rule element. This would result in four 'A's in the list
>> compared to the two for the wildcard.
>>
>> And there is also the difference that the execution of the action is
>> decoupled from the specific rule elements in the conjunct element
>> (causing the initial problem with the label expression)
>>
>>> In any case it is our idea and the truth is that we do not know how to
>>> predict the impact.
>> I can totally understand that. I had to debug the implementation myself
>> quite a bit in order to get to know what is going on. That does not
>> happen too often to me concerning ruta.
>>
>> I am open to suggestions how this situation can be improved. I am not
>> sure yet what should/can be done.
>>
>> Best,
>>
>> Peter
>>
>>> regards
>>>
>>> 2017-03-13 10:56 GMT+01:00 Peter Klügl <peter.kluegl@averbis.com>:
>>>
>>>> Hi,
>>>>
>>>>
>>>> there is a conceptual problem when using label expressions in those
>>>> conjunct rules. The label expression stores the value of the matched
>>>> annotations even if the rule element does not match. It has to since the
>>>> matched annotation may be used to validate the conditions as you do.
>>>>
>>>> Maybe the value should be reset if the rule element does not match
>>>> correctly, but I have to investigate the consequences furhter before I
>>>> implement this change. Are there opinions?
>>>>
>>>>
>>>> If you change the rule to not use the label expression in the action,
>>>> there still seems to be a problem with the conjunct rules. e.g. like:
>>>>
>>>> Document{ -> CREATE(Detection, "anchors" = list)} <- {
>>>>         w1:CW{w1.ct=="A" -> ADD(list, CW)}
>>>>         %
>>>>         w2:CW{w2.ct=="B" -> ADD(list, CW)}
>>>>        %
>>>>        w3:CW{w3.ct=="B" -> ADD(list, CW)}
>>>>        %
>>>>        w4:CW{w4.ct=="A" -> ADD(list, CW)};
>>>>        };
>>>>
>>>>
>>>> This looks like a bug when validating the match itself depednent on the
>>>> element matches. I need to investigate that further...
>>>>
>>>>
>>>> I'll create a jira ticket for both.
>>>>
>>>>
>>>> Best,
>>>>
>>>>
>>>> Peter
>>>>
>>>>
>>>>
>>>> Am 09.03.2017 um 11:23 schrieb José Vicente Moyano Murillo:
>>>>> Hi Peter
>>>>>
>>>>> We have prepared a short test with 2 methods.
>>>>> The first one is using #, the second one is using %
>>>>>
>>>>> We expect the same behaviour. But as you will see there are diferences.
>>>>>
>>>>> Regards
>>>>>
>>>>> 2017-02-28 16:00 GMT+01:00 Peter Klügl <peter.kluegl@averbis.com
>>>>> <mailto:peter.kluegl@averbis.com>>:
>>>>>
>>>>>     hmmm ok, maybe my example for testing the rule was too simple.
>>>>>
>>>>>     I will create a more complex example and check the results
>>>>>
>>>>>
>>>>>     Peter
>>>>>
>>>>>
>>>>>     Am 28.02.2017 um 15:38 schrieb José Vicente Moyano Murillo:
>>>>>     > Hi Peter,
>>>>>     >
>>>>>     > I'm sorry but this example does not work properly. It seems
that
>>>>>     it is a
>>>>>     > problem regarding conjunction rules (%)
>>>>>     >
>>>>>     > This is our test case and result.
>>>>>     >
>>>>>     > We have a book with 5 attributes
>>>>>     > - name
>>>>>     > - author
>>>>>     > - pages
>>>>>     > - ISBN
>>>>>     > - Category
>>>>>     >
>>>>>     > The rule is validatin name and author.
>>>>>     >
>>>>>     > RUTA creates the annotation with the list but the list contains
>>>>>     20 times
>>>>>     > the las attribute (Category).
>>>>>     >
>>>>>     > If we use # it works perfectly. But in general we must use %
>>>>>     >
>>>>>     > May be this information will help you.
>>>>>     >
>>>>>     > Many thanks
>>>>>     >
>>>>>     >
>>>>>     >
>>>>>     > 2017-02-28 14:23 GMT+01:00 Peter Kluegl <pkluegl@gmail.com
>>>>>     <mailto:pkluegl@gmail.com>>:
>>>>>     >
>>>>>     >> Hi,
>>>>>     >>
>>>>>     >>
>>>>>     >> operations directly on lists and arrays are on my todo list
but
>>>>>     not yet
>>>>>     >> implemented.
>>>>>     >>
>>>>>     >>
>>>>>     >> Right now, there are still some options like a variable
or
>>>>>     restricted
>>>>>     >> assignments.
>>>>>     >>
>>>>>     >>
>>>>>     >> Here's an example with an ANNOTATIONLIST variable:
>>>>>     >>
>>>>>     >>
>>>>>     >> ANNOTATIONLIST list;
>>>>>     >> Book{-> CREATE(NeilsBook, "attributes" = list)}<-{
>>>>>     >>     a1:Attribute{a1.name <http://a1.name>=="title",
>>>>>     a1.ct=="Norse Mythology" ->
>>>>>     >> ADD(list,a1)}
>>>>>     >>     % a2:Attribute{a2.name <http://a2.name>=="author",
>>>>>     a2.ct=="Neil Gaiman" ->
>>>>>     >> ADD(list,a2)};
>>>>>     >>     };
>>>>>     >>
>>>>>     >>
>>>>>     >> Variables are global, so you need to clear the list maybe
>>>>>     somewhere,
>>>>>     >> e.g. before the first ADD.
>>>>>     >>
>>>>>     >>
>>>>>     >> In the current trunk, something like the following is also
>>>>>     possible:
>>>>>     >>
>>>>>     >> Book{-> NeilsBook("attributes" = Attribute{OR(Attribute
>>>>>     >> .name=="title", Attribute.name=="author")})}....
>>>>>     >>
>>>>>     >>
>>>>>     >> Best,
>>>>>     >>
>>>>>     >>
>>>>>     >> Peter
>>>>>     >>
>>>>>     >>
>>>>>     >> Am 28.02.2017 um 14:06 schrieb José Vicente Moyano Murillo:
>>>>>     >>> Hi Peter, we have another question regarding the same
example.
>>>>>     >>>
>>>>>     >>> Right now all is working properly (many thanks) and
it seems
>> that
>>>>>     >>> 2.5.1-SNAPSHOT it's fine.
>>>>>     >>>
>>>>>     >>> Now we are planning to store some information in the
created
>>>>>     annotation.
>>>>>     >> We
>>>>>     >>> want to store all the attributes that matches the RUTA
rule.
>>>>>     >>>
>>>>>     >>> It is possible to store a1 and a2 in a feature (attributes)
of
>>>>>     NeilsBook?
>>>>>     >>>
>>>>>     >>> The example:
>>>>>     >>>
>>>>>     >>> Book{-> NeilsBook}<-{
>>>>>     >>>     a1:Attribute{a1.name <http://a1.name>=="title",
>>>>>     a1.ct=="Norse Mythology"}
>>>>>     >>>     %
>>>>>     >>> a2:Attribute{a2.name <http://a2.name>=="author",
a2.ct=="Neil
>>>>>     Gaiman"};
>>>>>     >>> };
>>>>>     >>>
>>>>>     >>> We are using this aproach:
>>>>>     >>>
>>>>>     >>> Book{-> NeilsBook("attributes" = Attribute )}<-{
>>>>>     >>>      a1:Attribute{a1.name <http://a1.name>=="title",
>>>>>     a1.ct=="Norse Mythology"}
>>>>>     >>>      %
>>>>>     >>> a2:Attribute{a2.name <http://a2.name>=="author",
a2.ct=="Neil
>>>>>     Gaiman"};
>>>>>     >>> };
>>>>>     >>>
>>>>>     >>> This example works but it is storing all the annotations
>>>>>     "Attribute"
>>>>>     >>> covered by "Book" and we just want the attributes that
matches
>>>>>     the rule.
>>>>>     >>>
>>>>>     >>> Thanks in advance.
>>>>>     >>>
>>>>>     >>>
>>>>>     >>>
>>>>>     >>>
>>>>>     >>> 2017-02-24 10:29 GMT+01:00 José Vicente Moyano Murillo
>>>>>     <moyanoj@aia.es <mailto:moyanoj@aia.es>>:
>>>>>     >>>
>>>>>     >>>> Many thanks Peter ¡¡¡
>>>>>     >>>>
>>>>>     >>>> 2017-02-24 9:46 GMT+01:00 Peter Klügl
>>>>>     <peter.kluegl@averbis.com <mailto:peter.kluegl@averbis.com>>:
>>>>>     >>>>
>>>>>     >>>>> Hi,
>>>>>     >>>>>
>>>>>     >>>>>
>>>>>     >>>>> the repo:
>>>>>     >>>>>
>>>>>     >>>>> <repository>
>>>>>     >>>>>       <id>apache.snapshots</id>
>>>>>     >>>>>       <name>Apache Snapshot Repository</name>
>>>>>     >>>>>       <url>http://repository.apache.org/snapshots
>>>>>     <http://repository.apache.org/snapshots></url>
>>>>>     >>>>>       <releases>
>>>>>     >>>>>         <enabled>false</enabled>
>>>>>     >>>>>       </releases>
>>>>>     >>>>>     </repository>
>>>>>     >>>>>
>>>>>     >>>>>
>>>>>     >>>>> the folder:
>>>>>     >>>>>
>>>>>     >>>>> https://repository.apache.org/content/groups/snapshots/org/a
>>>>>     <https://repository.apache.org/content/groups/snapshots/org/a>
>>>>>     >>>>> pache/uima/ruta-core/2.5.1-SNAPSHOT/
>>>>>     >>>>>
>>>>>     >>>>>
>>>>>     >>>>> Best,
>>>>>     >>>>>
>>>>>     >>>>> Peter
>>>>>     >>>>>
>>>>>     >>>>>
>>>>>     >>>>>
>>>>>     >>>>> Am 23.02.2017 um 13:47 schrieb José Vicente
Moyano Murillo:
>>>>>     >>>>>> Hi,
>>>>>     >>>>>>
>>>>>     >>>>>> ruta-core.jar it's enough.
>>>>>     >>>>>>
>>>>>     >>>>>> It will be fantastic if i could access the
snapshot
>>>> repository.
>>>>>     >>>>>>
>>>>>     >>>>>> 2017-02-23 13:03 GMT+01:00 Peter Klügl
>>>>>     <peter.kluegl@averbis.com <mailto:peter.kluegl@averbis.com>>:
>>>>>     >>>>>>
>>>>>     >>>>>>> Hi,
>>>>>     >>>>>>>
>>>>>     >>>>>>>
>>>>>     >>>>>>> what do you need? Only ruta-core.jar
or also the Eclipse
>>>>>     >> plugins/update
>>>>>     >>>>>>> site?
>>>>>     >>>>>>>
>>>>>     >>>>>>> I will prepare a new RC for the next
release soon.
>>>>>     >>>>>>>
>>>>>     >>>>>>> There should be snapshot artifacts built
by jenkins in the
>>>>>     snapshot
>>>>>     >>>>>>> repository. I am out-of-office today
so do not have the
>>>>>     link right
>>>>>     >>>>> now. (If
>>>>>     >>>>>>> you want to build it yourself, the svn
source repo can be
>>>>>     accessed by
>>>>>     >>>>>>> anyone)
>>>>>     >>>>>>>
>>>>>     >>>>>>>
>>>>>     >>>>>>> Best,
>>>>>     >>>>>>>
>>>>>     >>>>>>>
>>>>>     >>>>>>> Peter
>>>>>     >>>>>>>
>>>>>     >>>>>>>
>>>>>     >>>>>>>
>>>>>     >>>>>>>
>>>>>     >>>>>>> Am 23.02.2017 um 08:06 schrieb José
Vicente Moyano Murillo:
>>>>>     >>>>>>>
>>>>>     >>>>>>>> Hi Peter, good morning.
>>>>>     >>>>>>>>
>>>>>     >>>>>>>> Have we any opportunity to get a
fixed version? I mean a
>>>>>     snapshot or
>>>>>     >>>>> an
>>>>>     >>>>>>>> access to the repo.
>>>>>     >>>>>>>>
>>>>>     >>>>>>>> Regards
>>>>>     >>>>>>>>
>>>>>     >>>>>>>> 2017-02-22 10:38 GMT+01:00 José
Vicente Moyano Murillo <
>>>>>     >>>>> moyanoj@aia.es <mailto:moyanoj@aia.es>>:
>>>>>     >>>>>>>> you're right
>>>>>     >>>>>>>>> % is performing some kind of
"or".
>>>>>     >>>>>>>>>
>>>>>     >>>>>>>>> So we will wait for the new
release.
>>>>>     >>>>>>>>>
>>>>>     >>>>>>>>> Many thanks
>>>>>     >>>>>>>>>
>>>>>     >>>>>>>>> 2017-02-22 9:54 GMT+01:00 Peter
Klügl
>>>>>     <peter.kluegl@averbis.com <mailto:peter.kluegl@averbis.com>>:
>>>>>     >>>>>>>>>
>>>>>     >>>>>>>>> I actually wonder why your rules
work. I am quite sure
>>>>>     that they
>>>>>     >>>>> match
>>>>>     >>>>>>>>>> too often, i. e. they match
also if only one of the
>>>>>     conjunct rule
>>>>>     >>>>>>>>>> elements match.
>>>>>     >>>>>>>>>>
>>>>>     >>>>>>>>>> You can simply try that
by using a wrong string in the
>>>>>     check like
>>>>>     >>>>> "NG"
>>>>>     >>>>>>>>>> instead of "Neil Gaiman"
>>>>>     >>>>>>>>>>
>>>>>     >>>>>>>>>> Anyways, the problem will
be fixed in a few minutes and
>>>>>     will be
>>>>>     >>>>> part of
>>>>>     >>>>>>>>>> the next release.
>>>>>     >>>>>>>>>>
>>>>>     >>>>>>>>>>
>>>>>     >>>>>>>>>> Best,
>>>>>     >>>>>>>>>>
>>>>>     >>>>>>>>>>
>>>>>     >>>>>>>>>> Peter
>>>>>     >>>>>>>>>>
>>>>>     >>>>>>>>>>
>>>>>     >>>>>>>>>> Am 22.02.2017 um 09:38 schrieb
Peter Klügl:
>>>>>     >>>>>>>>>>
>>>>>     >>>>>>>>>>> Thanks. Maybe its just
a bug in 2.5.0 I already fixed.
>>>>>     I'll
>>>>>     >>>>> investigate
>>>>>     >>>>>>>>>> it.
>>>>>     >>>>>>>>>>
>>>>>     >>>>>>>>>>> Am 22.02.2017 um 09:24
schrieb José Vicente Moyano
>>>>>     Murillo:
>>>>>     >>>>>>>>>>>
>>>>>     >>>>>>>>>>>> Thank you very much
Peter. Your advice was amazing.
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>>> We tried the first
option using Conjunct rules and as
>>>>>     you said
>>>>>     >> it
>>>>>     >>>>> does
>>>>>     >>>>>>>>>>> not
>>>>>     >>>>>>>>>>> work with version 2.5.0.
But we change a little your
>>>>>     example and
>>>>>     >> it
>>>>>     >>>>>>>>>>> works
>>>>>     >>>>>>>>>>> perfectly witn 2.4.0
and 2.5.0
>>>>>     >>>>>>>>>>>> We use theses examples
with success:
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>>> DECLARE Annotation
RuleDetection;
>>>>>     >>>>>>>>>>>>   Book{ -> CREATE(NeilsBook)
} <- {
>>>>>     >>>>>>>>>>>>        Attribute{Attribute.name=="title",
>>>>>     Attribute.ct=="Norse
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>> Mythology"}"
>>>>>     >>>>>>>>>>>        %
>>>>>     >>>>>>>>>>>>                Attribute{Attribute.name=="author",
>>>>>     >>>>>>>>>>>> Attribute.ct=="Neil
>>>>>     >>>>>>>>>>>> Gaiman"};"
>>>>>     >>>>>>>>>>>>   };
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>>> DECLARE Annotation
RuleDetection;
>>>>>     >>>>>>>>>>>>   Book{ -> CREATE(NeilsBook)
} <- {
>>>>>     >>>>>>>>>>>>         Attribute{FEATURE("name","title"),
>>>>>     FEATURE("ct", "Norse
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>> Mythology")}"
>>>>>     >>>>>>>>>>>        %
>>>>>     >>>>>>>>>>>>                
Attribute{FEATURE("name","author"),
>>>>>     >> FEATURE("ct",
>>>>>     >>>>>>>>>>>> "Neil
>>>>>     >>>>>>>>>>>> Gaiman")};
>>>>>     >>>>>>>>>>>>   };
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>>> DECLARE Annotation
RuleDetection;
>>>>>     >>>>>>>>>>>>   Book{ -> CREATE(NeilsBook)
} <- {
>>>>>     >>>>>>>>>>>>        Attribute{Attribute.name=="title",
>>>>>     Attribute.ct=="Norse
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>> Mythology"}"
>>>>>     >>>>>>>>>>>        %
>>>>>     >>>>>>>>>>>>                
Attribute{FEATURE("name","author"),
>>>>>     >> FEATURE("ct",
>>>>>     >>>>>>>>>>>> "Neil
>>>>>     >>>>>>>>>>>> Gaiman")};
>>>>>     >>>>>>>>>>>>   };
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>>> May be the problem
is with when we use Identifiers:
>>>>>     >>>>>>>>>>>>     a1:Attribute
and  a2:Attribute
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>>> In any case thank
you very much for your help.
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>
>>>>>     >>>>>>>>>>>> 2017-02-21 17:46
GMT+01:00 Peter Klügl <
>>>>>     >> peter.kluegl@averbis.com <mailto:peter.kluegl@averbis.com>
>>>>>     >>>>>> :
>>>>>     >>>>>>>>>>>> Hi,
>>>>>     >>>>>>>>>>>>> I'd normally
say that you need the conjunt rules
>>>>>     construct to
>>>>>     >>>>> specify
>>>>>     >>>>>>>>>>>> an
>>>>>     >>>>>>>>>>> AND between two rule
element independent of the
>> position:
>>>>>     >>>>>>>>>>>>> Book{-> NeilsBook}<-{
>>>>>     >>>>>>>>>>>>>      a1:Attribute{a1.name
<http://a1.name>=="title",
>>>>>     a1.ct=="Norse Mythology"}
>>>>>     >>>>>>>>>>>>>      % a2:Attribute{a2.name
>>>>>     <http://a2.name>=="author", a2.ct=="Neil Gaiman"};
>>>>>     >>>>>>>>>>>>>      };
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>> However, I just
noted that there is a problem with
>>>>>     conjunct
>>>>>     >>>>> rules. I
>>>>>     >>>>>>>>>>>>> haven't used
it in a long time and the test coverage
>>>>>     much lower
>>>>>     >>>>> than
>>>>>     >>>>>>>>>>>> the
>>>>>     >>>>>>>>>>> other constructs. I'll
create a ticket for it and fix
>> it.
>>>>>     >>>>>>>>>>>>> Without conjunct
rules, you need some boolean
>>>>>     variables for
>>>>>     >>>>> cheking
>>>>>     >>>>>>>>>>>> the
>>>>>     >>>>>>>>>>> AND, which looks all
but declarative:
>>>>>     >>>>>>>>>>>>> BOOLEAN ft,
fa;
>>>>>     >>>>>>>>>>>>> FOREACH(book)
Book{}{
>>>>>     >>>>>>>>>>>>>      book{->
ft = false, fa = false};
>>>>>     >>>>>>>>>>>>>      book->{a1:Attribute{a1.name
>>>>>     <http://a1.name>=="title", a1.ct=="Norse
>>>>>     >>>>>>>>>>>>> Mythology"->
>>>>>     >>>>>>>>>>>>> ft=true};};
>>>>>     >>>>>>>>>>>>>      book->{a2:Attribute{a2.name
>>>>>     <http://a2.name>=="author", a2.ct=="Neil
>>>>>     >>>>> Gaiman"->
>>>>>     >>>>>>>>>>>>> fa=true};};
>>>>>     >>>>>>>>>>>>>      book{ft,fa
-> NeilsBook};
>>>>>     >>>>>>>>>>>>> }
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>> ... or with
a BLOCK...
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>> BLOCK(book)
Book{}{
>>>>>     >>>>>>>>>>>>>      Document{->
ft = false, fa = false};
>>>>>     >>>>>>>>>>>>>      a1:Attribute{a1.name
<http://a1.name>=="title",
>>>>>     a1.ct=="Norse Mythology"->
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>> ft=true};
>>>>>     >>>>>>>>>>>      a2:Attribute{a2.name
<http://a2.name>=="author",
>>>>>     a2.ct=="Neil Gaiman"->
>>>>>     >>>>> fa=true};
>>>>>     >>>>>>>>>>>>>      Document{ft,fa
-> NeilsBook};
>>>>>     >>>>>>>>>>>>> }
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>> If the order
of the attributes is known, you can
>>>>>     avoid the AND
>>>>>     >>>>> check
>>>>>     >>>>>>>>>>>> and
>>>>>     >>>>>>>>>>> just specify a sequential
constraint:
>>>>>     >>>>>>>>>>>>> Book{-> NeilsBook}<-{
>>>>>     >>>>>>>>>>>>>      a1:Attribute{a1.name
<http://a1.name>=="title",
>>>>>     a1.ct=="Norse Mythology"}
>>>>>     >>>>>>>>>>>>>      # a2:Attribute{a2.name
>>>>>     <http://a2.name>=="author", a2.ct=="Neil Gaiman"};
>>>>>     >>>>>>>>>>>>>      };
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>> If you need
to check on the complete title with the
>>>>>     URL, you
>>>>>     >> can
>>>>>     >>>>>>>>>>>> replace
>>>>>     >>>>>>>>>>> the coveredText comparison
with a REGEXP condition.
>>>>>     >>>>>>>>>>>>> (tested with
UIMA Ruta 2.5.0)
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>> Best,
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>> Peter
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>> Am 21.02.2017
um 13:58 schrieb José Vicente Moyano
>>>>>     Murillo:
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>> Hello everyone
>>>>>     >>>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>> I'm planning
to use RUTA to create some
>>>>>     annotations. But i'm
>>>>>     >> not
>>>>>     >>>>>>>>>>>>> able to
>>>>>     >>>>>>>>>>> accomplish my objective.
>>>>>     >>>>>>>>>>>>>> This is
my case right now:
>>>>>     >>>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>> I have a
text annotated with some annotations
>>>> "*Book*".
>>>>>     >>>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>> Under "*Book*"
annotation i have a few annotations
>>>>>     >> "*Attribute*"
>>>>>     >>>>>>>>>>>>>> that
>>>>>     >>>>>>>>>>>>>> stores some
information about the book. Each
>>>>>     "*Attribute*"
>>>>>     >> has a
>>>>>     >>>>>>>>>>>>> feature
>>>>>     >>>>>>>>>>> "
>>>>>     >>>>>>>>>>>>>> *name*"
and a feature "*parent*" to its parent (one
>>>>>     "*Book*"
>>>>>     >>>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>> annotation).
>>>>>     >>>>>>>>>>> And example could be
a text with 2 "*Book*"
>> annotations:
>>>>>     >>>>>>>>>>>>>> "*Book*"
annotation number 1 with 3 nested
>> attributes
>>>>>     >>>>>>>>>>>>>> 1.- *Attribute*
with name feature equals "title" ->
>>>>>     covered
>>>>>     >>>>> text:
>>>>>     >>>>>>>>>>>>> "Norse
>>>>>     >>>>>>>>>>> Mythology
>>>>>     >>>>>>>>>>>>>>
>>>>>     <http://www.barnesandnoble.com/w/norse-mythology-neil-
>>>>>     <http://www.barnesandnoble.com/w/norse-mythology-neil->
>>>>>     >>>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>
>>>>>     gaiman/1124023596;jsessionid=FD1D8F9690602616CA59B38CFE9290
>>>>>     >>>>>>>>>>>>> 06.prodny_store02-atgap08?ean=9780393609097>
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>> "
>>>>>     >>>>>>>>>>>>>> 2.- *Attribute*
with name feature equals "author"
>>>>>     ->  covered
>>>>>     >>>>> text:
>>>>>     >>>>>>>>>>>>> "Neil
>>>>>     >>>>>>>>>>> Gaiman"
>>>>>     >>>>>>>>>>>>>> 3.- *Attribute*
with name feature equals "language"
>> -
>>>>>     >> covered
>>>>>     >>>>>>>>>>>>>> text:
>>>>>     >>>>>>>>>>>>>> "English"
>>>>>     >>>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>> "*Book*"
annotation number 2 with 3 nested
>> attributes
>>>>>     >>>>>>>>>>>>>> 1.- *Attribute*
with name feature equals "title" ->
>>>>>     covered
>>>>>     >>>>> text:
>>>>>     >>>>>>>>>>>>> "Never
>>>>>     >>>>>>>>>>> Never
>>>>>     >>>>>>>>>>>>>>
>>>>>     <http://www.barnesandnoble.com/w/never-never-james-patterson
>>>>>     <http://www.barnesandnoble.com/w/never-never-james-patterson>
>>>>>     >>>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>> /1123863634;
>>>>>     >>>>>>>>>>>
>>>>>     jsessionid=FD1D8F9690602616CA59B38CFE929006.prodny_store02-
>>>>>     >>>>>>>>>>>> atgap08?ean=
>>>>>     >>>>>>>>>>> 9780316433174>
>>>>>     >>>>>>>>>>>>>> "
>>>>>     >>>>>>>>>>>>>> 2.- *Attribute*
with name feature equals "author"
>>>>>     ->  covered
>>>>>     >>>>> text:
>>>>>     >>>>>>>>>>>>> "James
>>>>>     >>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>> Patterson"
>>>>>     >>>>>>>>>>>>>> 3.- *Attribute*
with name feature equals "language"
>> -
>>>>>     >> covered
>>>>>     >>>>>>>>>>>>>> text:
>>>>>     >>>>>>>>>>>>>> "English"
>>>>>     >>>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>> I need to
respect this schema but i have this
>>>> question:
>>>>>     >>>>>>>>>>>>>> It is possible
to create and annotation over a book
>>>>>     for a
>>>>>     >> given
>>>>>     >>>>>>>>>>>>> author
>>>>>     >>>>>>>>>>> name
>>>>>     >>>>>>>>>>>>>> and a given
title name?
>>>>>     >>>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>> Thank's
in advance
>>>>>     >>>>>>>>>>>>>>
>>>>>     >>>>>>>>>>>>>>
>>>>>     >>
>>>>>
>>>>>
>>


Mime
View raw message