Ling 571 - Deep Processing Techniques for NLP
Winter 2016
Homework #1: Due January 12, 2016: 11:45pm

Goals

• Explore the basics of context-free grammar design.
• Identify some of the challenges in building grammars for natural languages.
• Begin to gain some familiarity with the Natural Language Toolkit (NLTK).
• Gain some experience with the cluster and condor.

Background

Building a Grammar

Your grammar should be able to produce parses for all sentences in the files (as well as other similar ones in the English language). The grammar should capture the major clause type (S, etc.), the major phrase types (NP, VP, PP, etc.), the parts of speech (POS) (NN, VBZ), and any punctuation or special symbols. The phrase and POS types specified in the Jurafsky and Martin text (CH. 12 and inside front cover) provide a good basis for your grammar.

You may hard-code capitalization.

Parsing

• Load your grammar.
• Build a parser for your grammar using nltk.parse.EarleyChartParser.
• Read in the example sentences.
• For each example sentence, output to a file
• the sentence itself
• the simple bracketed structure parse(s), and
• the number of parses for that sentence.
• Finally, print the average number of parses per sentence obtained by your grammar.

Programming

• <grammar_file> is the name of the file holding your grammar rules in the NLTK .cfg format.
• <test_sentence_file> is the name of the file holding the set of sentences to parse, one sentence per line
• <output_file> is the name of output file for your system

Files

Test and Example Files

• sentences.txt: Sentences to test against
• toy.cfg: Toy example NLTK-format grammar file. Other examples can be found on patas under /corpora/nltk/nltk-data/grammars/
• toy_sentences.txt: Example set of sentences to parse with toy grammar
• toy_output.txt: Example output file based on toy grammar

Submission files

• hw1_parse.py: Primary program file
• hw1_grammar.cfg: Your grammar file
• hw1_output.txt: Results of running your parsing program on the test sentences with your grammar.
• hw1.cmd: Condor file which drives your parser.
  • Your program must run on patas using:
    $ condor_submit hw1.cmd
    Please see the CLMS wiki pages on the basics of using the condor cluster.

    All files created by the condor run should appear in the top level of the directory.

• readme.{txt|pdf}: Write-up file
  • This file should describe and discuss your work on this assignment. Include problems you came across and how (or if) you were able to solve them, any insights, special features, and what you learned. Give examples if possible. If you were not able to complete parts of the project, discuss what you tried and/or what did not work.
• hw1.tar: Your hand-in file
  • Use the tar command to build a single hand-in file, named hw#.tar where # is the number of the homework assignment and containing all the material necessary to test your assignment. Your hw1.cmd should be at the top level of whatever directory structure you are using.

    For example, in your top-level directory, run:
    $ tar cvf hw1.tar *

Handing in your work