Untitled
University of Toronto Computer S
ien
e 485/2501
St. George Campus Autumn 2018
CSC 485/2501: Introdu
tion to Computational Linguisti
s
Course Information
Instru
tor : Gerald Penn
Le
tures: R 11{1, WI 1016
OÆ
e: PT 396B
Tel: (416)978-7390
OÆ
e Hours: immediately pre
eding le
tures 10{11, or by appointment
Email: gpenn�tea
h.
s.utoronto.
a
Tutorials: F 2{3, WI 1016
(Note: some tutorial days will be used for le
tures)
Tea
hing Assistants: Name Assignment
Sean Robertson 1
Nona Naderi 2
Aditya Bhargava 3
Textbooks:
Required Jurafsky, Daniel, and Martin, James H. Spee
h and Language Pro
essing ,
2nd edition, Pearson Prenti
e-Hall, 2009. Available in paper and e-book
rental versions (for the latter, go to CourseSmart.
om and sear
h for
Jurafsky). We’ll also be referring to the draft 3rd edition:
https://web.stanford.edu/�jurafsky/slp3/. See also the errata list
for the 2nd edition: www.
s.
olorado.edu/�martin/SLP/Errata/
SLP2-PH-Errata.html.
Required Bird, Steven; Klein, Ewan; and Loper, Edward. Natural Language
Pro
essing with Python, O’Reilly, 2009. Free (in HTML) with online
extras at www.nltk.org/book.
Re
ommended Mertz, David. Text Pro
essing in Python. Addison-Wesley, 2003. Free ASCII
version at Gnosis.
x/TPiP.
Optional Allen, James. Natural Language Understanding , 2nd edition. Benjamin/
Cummings, 1995.
Re
ommended Martelli, Ravens
roft and Holden. Python in a Nutshell, 3rd ed., O’Reilly, 2017.
Course Web Page : http://www.
s.toronto.edu/~gpenn/
s
485/
Evaluation : For undergraduates registered in CSC 485, there will be three homework assign-
ments worth one third of your �nal mark ea
h. Those registered for CSC 2501 must
omplete
the three homework assignments (25% ea
h), as well as �ve essays on assigned resear
h papers
(5� 5 = 25%). There is no �nal examination for either
ourse
ode.
� No late homeworks will be a
epted, ex
ept in
ase of do
umented medi
al or other emer-
gen
ies.
Poli
y on
ollaboration : No
ollaboration on homeworks or essays is permitted. The work
you submit must be your own.
Failure to observe this poli
y is an a
ademi
o�ense,
arrying a penalty ranging from a zero
on the homework to suspension from the university.
1
Course Goals: This
ourse is an introdu
tion to a statisti
al and
omputational
hara
teriza-
tion of natural language. You will also have the
han
e to pra
ti
e programming in Python.
Prerequisites: For undergraduates, STA247H1/STA255H1/STA257H1 and CSC209H1, but
CSC324H1/CSC330H1/CSC384H1 is strongly re
ommended. For advi
e,
onta
t the Under-
graduate OÆ
e on the fourth
oor of the Bahen Centre or the instru
tor.
Newsgroup: The
ourse newsgroup is on the web at
https://bb-2018-09.tea
h.
s.toronto.edu/
/
s
485. Your tea
hing assistants will be
monitoring it.
Tentative Syllabus:
Date Topi
Advan
e reading*
6{13 Sept Intro to
omputational linguisti
s RP; J&M: 1; BK&L: 1, 2.3, 4 as ne
essary
13{20 Sept Grammars and parsing J&M: 5.0{1, 12.0{12.3.3, 12.3.7,
13.1{2; BK&L: 8.0{8.4
21 Sept Intro to NLTK BK&L: 1, 2.3, 4 as ne
essary
27{28 Sept Chart parsing RP; J&M: 13.3{4; A: 3.4, 3.6; BK&L: 8.4 and
online extras se
tion 8.2 on
hart parsing
4{12 O
t Even More Parsing RP (5th); J&M: 12.3.4{6, 15.0{3; A: 4.1{5;
BK&L: 9
12{18 O
t Ambiguity resolution
18{25 O
t Statisti
al atta
hment disambiguation RP
25 O
t { Lexi
al semanti
s J&M: 19.1{4, 20.8
1 Nov
1{2 Nov Word sense disambiguation RP; J&M: 20.1{5
15{16 Nov Statisti
al parsing J&M: 5.2{5.5.2, 5.6, 12.4, 14.0{1, 14.3{7
16{22 Nov Anaphora resolution 21.0, 21.2{8
23{29 Nov Semanti
representations J&M: 17.0{17.4.1, 17.5; BK&L: 10.0{4
30 Nov Tutorial will meet as usual
*J&M = Jurafsky and Martin; BK&L = Bird, Klein, and Loper; A = Allen; RP = resear
h
paper distributed on-line; itali
s indi
ates optional additional reading.
2
Tentative Course Calendar :
Thu, 6 September First le
ture
Thu, 13 September Write-up 1 due (CSC 2501)
Wed, 19 September Last day to add
ourse (CSC 485)
Mon, 24 September Last day to add
ourse (CSC 2501)
Thu, 27 September Write-up 2 due (CSC 2501)
Fri, 5 O
tober Assignment 1 due
Thu, 11 O
tober Write-up 3 due (CSC 2501)
Thu, 25 O
tober Write-up 4 due (CSC 2501)
Mon, 29 O
tober Last day to drop
ourse (CSC 2501)
Thu, 1 November Assignment 2 due
Mon, 5 November Last day to drop
ourse (CSC 485)
5{9 November Reading Week | no le
tures or tutorial
Thu, 15 November Write-up 5 due (CSC 2501)
Thu, 29 November Last le
ture
Thu, 6 De
ember Assignment 3 due
3