Grammar Writing: Week 10: Difference between revisions

From English Grammar
Jump to navigation Jump to search
(Created page with "= Identity = Last week, you were supposed to write a grammar that accepts different kinds of verb phrases. The rules you had to write all took the following form: # VP --> V...")
 
 
(14 intermediate revisions by 2 users not shown)
Line 3: Line 3:
Last week, you were supposed to write a grammar that accepts different kinds of verb phrases. The rules you had to write all took the following form:
Last week, you were supposed to write a grammar that accepts different kinds of verb phrases. The rules you had to write all took the following form:


# VP --> V, COMPS <NP> + NP
1. VP --> V, COMPS <NP> + NP<br>
# VP --> V, COMPS <PP> + PP
2. VP --> V, COMPS <PP> + PP<br>
# VP --> V, COMPS <AP> + AP
3. VP --> V, COMPS <AP> + AP<br>
# VP --> V, COMPS <S> + S
4. VP --> V, COMPS <sentence> + sentence<br>


For NPs, APs, PPs you would also have to write several rules.
For NPs, APs, PPs you would also have to write several rules.
Line 12: Line 12:
Clearly, the rules above are very similar, but they fail to express the similarity. This is shown by the fact that if there were a verb that wants to combine with a VP-complement, then we would would need to add another rule to the ones above, as follows:
Clearly, the rules above are very similar, but they fail to express the similarity. This is shown by the fact that if there were a verb that wants to combine with a VP-complement, then we would would need to add another rule to the ones above, as follows:


# VP --> V, COMPS <VP> + VP
5. VP --> V, COMPS <VP> + VP


What our current grammars are missing to express the generalization covered by the five rules above is the concept of '''identity'''.
What our current grammars are missing to express the generalization covered by the five rules above is the concept of '''feature structure identity'''. By identity, we mean that one larger feature structure contains a smaller feature structure twice, in different places. Or more simply, that the information at two places in a feature structure must be the same.
 
In order to signal such identity, we will use so-called boxed numbers, i.e. 1,2,3, ... with a box around it. Unfortunately, in this WIKI, we cannot represent such boxes. Therefore, we will use symbols like the following as identity markers: &copy;1, &copy;2, &copy;3, ...
 
With the concept of feature structure identity and the notation above, we can now replace the 5 rules above by the one single rule below:
 
6. VP --> V, COMPS <&copy;1> + &copy;1
 
The effect of rule 6. can be stated informally as follows:
 
''A verb phrase can be formed from a V that is looking for a &copy;1 as its complement and one &copy;1.''
 
 
<br>
 
= Homework (due date: Sunday, 8 p.m. on January 7th 2018) =
 
In this homework assignment, you are supposed to implement a very simple grammar. It should recognize the following four sentences.
 
 
----
 
# trump defied bureaucracy.[https://www.nytimes.com/2017/12/23/us/politics/trump-immigration.html?rref=collection%2Ftimestopic%2FTrump%2C%20Donald%20J.]
# trump likes putin.
# trump is happy.
# putin believes trump is happy.
 
----
 
 
Crucially, however, in this exercise you should use no more than '''two''' rules. Note that, previously, we had a classical sentence rule, plus one for a finite verb and its object, a rule for a finite verb which wants to combine with an adjective, one for sentential complement, etc.
 
The classical sentence rule is still needed, but the new rule '''verb_rule''' should now match all of the following patterns (from above):
 
 
VP --> V, COMPS <NP> + NP              // trump likes putin <br>
VP --> V, COMPS <AP> + AP              // trump is happy <br>
VP --> V, COMPS <sentence> + sentence  // putin believes trump is happy <br>
 
As an illustration, consider the following example.
 
 
 
[[File:rule1.JPG]]
 
The rule indicates identity relations between part of speech categories of the first daughter and the mother node, as well as shared information on the item in the complement list, which is structurally identical to the second daughter's SYNTAX information. Ideally, this rule should capture all above cases (... in fact, it does -- I've tried it ;-)).
 
Your task:
Implement both rules and upload your grammar. This exercise will be graded. You should not attempt to upload other people's zip files (unless you know how to modify the meta data of the original files.. ;-))
In case you're having trouble... nschenk@em.uni-frankfurt.de
 
 
 
 
 
<br>
<br>
Navigation:
<div align="center">
  [[Grammar_Writing|'''Main Page''']] &nbsp; [[Grammar_Writing:_Week_2| '''Week 2''']] &nbsp;&nbsp; [[Grammar_Writing:_Week_3| '''Week 3''']] &nbsp;&nbsp; [[Grammar_Writing:_Week_4| '''Week 4''']] &nbsp;&nbsp; [[Grammar_Writing:_Week_5| '''Week 5''']] &nbsp;&nbsp; [[Grammar_Writing:_Week_6| '''Week 6''']] &nbsp;&nbsp; [[Grammar_Writing:_Week_7| '''Week 7''']] &nbsp; [[Grammar_Writing:_Week_8| '''Week 8''']] &nbsp;&nbsp;  [[Grammar_Writing:_Week_9| '''Week 9''']] &nbsp;&nbsp; '''Week 10''' &nbsp;&nbsp;  [[Grammar_Writing:_Week_11| '''Week 11''']]

Latest revision as of 11:49, 8 January 2018

Identity

Last week, you were supposed to write a grammar that accepts different kinds of verb phrases. The rules you had to write all took the following form:

1. VP --> V, COMPS <NP> + NP
2. VP --> V, COMPS <PP> + PP
3. VP --> V, COMPS <AP> + AP
4. VP --> V, COMPS <sentence> + sentence

For NPs, APs, PPs you would also have to write several rules.

Clearly, the rules above are very similar, but they fail to express the similarity. This is shown by the fact that if there were a verb that wants to combine with a VP-complement, then we would would need to add another rule to the ones above, as follows:

5. VP --> V, COMPS <VP> + VP

What our current grammars are missing to express the generalization covered by the five rules above is the concept of feature structure identity. By identity, we mean that one larger feature structure contains a smaller feature structure twice, in different places. Or more simply, that the information at two places in a feature structure must be the same.

In order to signal such identity, we will use so-called boxed numbers, i.e. 1,2,3, ... with a box around it. Unfortunately, in this WIKI, we cannot represent such boxes. Therefore, we will use symbols like the following as identity markers: ©1, ©2, ©3, ...

With the concept of feature structure identity and the notation above, we can now replace the 5 rules above by the one single rule below:

6. VP --> V, COMPS <©1> + ©1

The effect of rule 6. can be stated informally as follows:

A verb phrase can be formed from a V that is looking for a ©1 as its complement and one ©1.



Homework (due date: Sunday, 8 p.m. on January 7th 2018)

In this homework assignment, you are supposed to implement a very simple grammar. It should recognize the following four sentences.



  1. trump defied bureaucracy.[1]
  2. trump likes putin.
  3. trump is happy.
  4. putin believes trump is happy.


Crucially, however, in this exercise you should use no more than two rules. Note that, previously, we had a classical sentence rule, plus one for a finite verb and its object, a rule for a finite verb which wants to combine with an adjective, one for sentential complement, etc.

The classical sentence rule is still needed, but the new rule verb_rule should now match all of the following patterns (from above):


VP --> V, COMPS <NP> + NP // trump likes putin
VP --> V, COMPS <AP> + AP // trump is happy
VP --> V, COMPS <sentence> + sentence // putin believes trump is happy

As an illustration, consider the following example.


Rule1.JPG

The rule indicates identity relations between part of speech categories of the first daughter and the mother node, as well as shared information on the item in the complement list, which is structurally identical to the second daughter's SYNTAX information. Ideally, this rule should capture all above cases (... in fact, it does -- I've tried it ;-)).

Your task: Implement both rules and upload your grammar. This exercise will be graded. You should not attempt to upload other people's zip files (unless you know how to modify the meta data of the original files.. ;-)) In case you're having trouble... nschenk@em.uni-frankfurt.de





Navigation:

 Main Page    Week 2     Week 3     Week 4     Week 5     Week 6     Week 7    Week 8      Week 9    Week 10      Week 11