C661: Final Examination
Fall 1995
Instructions: Answer the first question (Part A)
and both
questions in one of the two
groups (B and C), the group which differs more from your project.
If you have questions about the exam, in particular if terms
are not clear, please post them to the newsgroup so
everyone will see the response.
The exam is due at 11:59:59 pm on Monday, December 11.
Either electronic or hardcopy submissions are acceptable.
A. Symbolic and Connectionist NLP
- Compare a symbolic and a connectionist approach
to script-based story understanding with respect to the following:
- pronoun reference resolution
- inference
- script instantiation
B. Symbolic NLP
- Show (a) how a bottom-up chart parser would handle the following
sentence containing a novel word and (b) how the resulting parse
might be used to infer the syntactic category of the novel word.
- John placed the chair zelig the table.
Your grammar rules need not be of the fancy HPSG variety.
- Show how a unification grammar would handle article-noun agreement
in English.
You will need a distinction between mass and count nouns (e.g., water,
love, honesty vs.
cup, word, idea) and between
singular count and plural count or mass
indefinite articles (a vs. some).
Your grammar should accept the following:
- a cup
- some cups
- some water
but reject the following:
- *some cup
- *a cups
- *a water
Give the necessary grammar rules and the lexical entries for
the words a, some, water, cup, and
cups,
and show how the phrases would succeed or fail during parsing.
You can ignore the details of the parser (assuming
that the right entries or rules are magically selected),
and your rules and lexical entries can deviate from standard HPSG
ones.
C. Connectionist and Statistical NLP
- You have the task of developing a system which can assign
one of a set of 10,000 classifications (e.g.,
medieval European history,
cell biology, knowledge representation) to the journal articles
received by a library.
The system should work without the benefit of a grammar or
a dictionary but simply make use of the statistical properties of the
texts.
Describe the system you design for this task and how you would
train it.
- What connectionist options are there for representing
(or learning to represent) the structure of phrases such as the following?
- a green square
- a green square to the right of a circle
- a green square to the right of a circle containing two
blue triangles
- a green square to the right of a circle containing two
blue triangles on either side of a green square
Take me back to the C661
Home Page.
Last updated: 8 December 1995, 0:03
URL: http://www.cs.indiana.edu/classes/c661/exam.html
Comments:
gasser@salsa.indiana.edu
Copyright
1995, The Trustees of
Indiana University