[SOLVED] CSC 485H/2501H: Computational linguistics, Fall 2021

University of Toronto, Department of Computer Science

CSC 485H/2501H: Computational linguistics, Fall 2021

Copyright By Assignmentchef assignmentchef

Assignment 3

Due date: 23:59 on Thursday, December 9, 2021.

Late assignments will not be accepted without a valid medical certificate or other documentation

of an emergency.

For CSC485 students, this assignment is worth 33% of your final grade.

For CSC2501 students, this assignment is worth 25% of your final grade.

Read the whole assignment carefully.

Type the written parts of your submission in no less than 12pt font.

What you turn in must be your own work. You may not work with anyone else on any of

the problems in this assignment. If you need assistance, contact the instructor or TA for the

assignment.

Any clarifications to the problems will be posted on the Piazza forum for the class. You will

be responsible for taking into account in your solutions any information that is posted there,

or discussed in class, so you should check the page regularly between now and the due date.

The starter code directory for this assignment is accessible on Teaching Labs machines at

the path /u/csc485h/fall/pub/a3/. In this handout, code files we refer to are located in

that directory.

When implementing code, make sure to read the docstrings as some of them provide im-

portant instructions, implementation details, or hints.

Fill in your name, student number, and UTORid on the relevant lines at the top of each

file that you submit. (Do not add new lines; just replace the NAME, NUMBER, and UTORid

placeholders.)

Overview: Symbolic Machine Translation

In this assignment, you will learn how to write phrase structure grammars for some different

linguistic phenomena in two different languages: English and Chinese. You can use the two gram-

mars to create an interlingual machine translation system by parsing in one, and generating in the

other. Dont panic if you dont speak Chinese, and also dont cheer up yet if you can speak the lan-

guage it wont give you much of an advantage over other students. A facility with languages in

general will help you, as will the ability to learn and understand the nuances between the grammars

of two different languages.

In particular, you will start by working on agreement. Then, you will need to analyse the quan-

tifier scoping difference between the two languages.

TRALE Instructions The TRALE system can be run with:

/h/u2/csc485h/fall/pub/trale/trale -fsug

(which you are welcome to alias). For this assignment, TRALE needs to start a graphical interface:

Gralej. Therefore, if you dont have access to the labs and want to run TRALE remotely, you can

either use:

RDP over SSH1,

Remote Access Server NX2,

or connect to teach.cs using ssh with either the -X or -Y flag:

1. Agreement: Determiners, Numbers and Classifiers [10 marks]

English expresses subjectverb agreement in person and number. English has two kinds of

number: singular and plural. The subject of a clause must agree with its predicate: they should be

both singular or both plural. However, the number of a direct object does not need to agree with

(1) A professor steals a cookie.

(2) Two professors steal a cookie.

(3) * Two professors steals two cookies.

1https://www.teach.cs.toronto.edu/using cdf/rdp.html
2https://www.teach.cs.toronto.edu/using cdf/remote access server.html

https://www.teach.cs.toronto.edu/using_cdf/rdp.html
https://www.teach.cs.toronto.edu/using_cdf/remote_access_server.html

(4) * A professor steal two cookies.

Chinese, on the other hand, does not exhibit subjectverb agreement. As shown in the examples

below, most nouns do not inflect at all for plurality. Chinese does, however, have a classifier (CL)

part of speech that English does not. Semantically, classifiers are similar to English collective nouns

(a bottle of water, a murder of crows), but English collective nouns are only used when describing

collectives. With very few exceptions, classifiers are mandatory in complex Chinese noun phrases.

Different CLs agree with different classes of nouns that are sorted by mostly semantic criteria.

For example, (jiaoshou) professor is a person and an occupation, so it should be classified

by either (ge) or (wei) and cannot be classified by the animal CL (zhi). However, the

rules of determining a nouns class constitute a formal system that must be followed irrespective

of semantic similarity judgements. For example, while cats and dogs are both pets and can both

be classified by the animal CL (zhi), (gou) dog can take another classifier, (tiao), for

string-like objects.

You should be familiar by now with the terminology in the English grammar starter code for

this question. The Chinese grammar is fairly similar, but there is a new phrasal category called a

classifier phrase (CLP), formed by a number and a classifier. The classifier phrase serves the same

role as a determiner does in English.

The two grammars below dont appropriately constrain the NPs generated. You need to design

your own rules and features to properly enforce agreement.

English Grammar:

NP Det N
NP Num N

three: Num

professor: N

professors: N

Chinese Grammar:

NP CLP N
CLP Num CL

yi one/a: Num

liang two: Num

san three: Num

mao cat: N

gou dog: N

jiaoshou professor: N

kanjian see: V

zhui chase: V

tiao: CL

Here is a list of all of the nouns in this question and their acceptable classifiers:

mao cat: zhi;

gou dog: zhi, tiao;

jiaoshou professor: ge, wei.

(a) (7 marks) Implement one grammar for each language pursuant to the specifications above.

English: q1_en.pl and Chinese: q1_zh.pl.

Neither of your grammars need to handle embedded clauses, e.g., a professor saw two cats

chase a dog. Similarly for Chinese, your grammar doesnt need to parse sentences like ex-

ample (15):

A professor saw two cats chase a dog.

For the Chinese grammar, the lexical entries can be coded in either pinyin (the Romanized

transcriptions of the Chinese characters) or in simplified Chinese characters.

(b) (2 marks) Use your grammars to parse and translate the following sentences. Save and submit

all the translation results in the .grale format. The results of sentence (16) should be named

q1b_en.grale and the results of sentence (17) should be named q1b_zh.grale.

(16) Two cats chase one dog

Operational Instructions

If you decide to use simplified Chinese characters, enter them in Unicode and use the

-u flag when you run TRALE.

Independently test your grammars in TRALE first, before trying to translate.

Use the function translate to generate a semantic representation of your source sen-

tence. If your sentence can be parsed, the function translate should open another

gralej interface with all of the translation results.

| ?- translate([two,cats,chase,one,dog]).

To save the translation results, on the top left of the Gralej window (the window with

the INITIAL CATEGORY entry and all of the translated sentences listed), click File >>

Save all >> TRALE format.

Dont forget to close all of the windows or kill both of the Gralej processes after you

finish. Each Gralej process will take up one port in the server, and no one can use the

server if we run out of ports.

(c) (1 mark) Compare your translator with Google Translate3. At its core, Google Translate is a

neural machine translation (NMT) system. In a few sentences, describe the similarities and

differences between Google Translate and your system. Your analysis should be submitted

as the section 1(c) in analysis.txt.

2. Quantifier Scope [30 marks]

Quantifiers For this assignment, we will consider two quantifiers: the universal quantifier ( ev-

ery, mei) and the existential quantifier (a, yi). In English, both quantifiers behave as singular

determiners.

(18) A professor stole every cookie.

(19) * A professor stole every cookies.

(20) * A professors stole every cookie.

In Chinese, both of these quantifiers behave more like numerical determiners. In addition, when

a universal quantifier modifies an NP that occurs before the verb (such as with a universally quanti-

fied subject), the preverbal operator (dou) is required. When a universally quantified NP occurs

after the verb, the dou-operator must not appear with it.

3https://translate.google.ca/

https://translate.google.ca/

(21) Every professor stole a cookie.

(22) A professor stole every cookie.

We shall simplify our analysis of NPs in this question to be a sequence of a quantifier, a classi-

fier and a noun, and forget all about other determiners such as numbers.

Quantifier Scope Ambiguity In lecture, we talked about different kinds of ambiguity. Quantifier

scope ambiguity was one of them. In many English sentences, no matter what the order of the

quantifiers, there is a quantifier scope ambiguity. For example, there can be two readings of this

sentence (27):

(> ) Every student read a book. The books title is The Old Man and the Sea.

(> ) Every student read a book. Some students read The Old Man and the Sea.

( > ) means the existential quantifier outscopes the universal quantifier in a logical form repre-
sentation of the sentence.

(27) Every

Ambiguous: > and >

Ambiguous: > and >

Ambiguous: > and >

Unambiguous: only >

The English sentences (27,29) have a scope ambiguity no matter what the order of the quan-

tifiers. In Chinese, however, the sentence is only ambiguous if the universal quantifier came first

Received a coded retreat message we have. Master Yoda

Topicalization and Movement Topicalization is a linguistic phenomenon in which an NP ap-

pears at the beginning of a sentence in order to establish it as the topic of discussion in a sentence

or to emphasize it in some other way. It plays an important role in the syntax of fixed-word-order

languages because grammatical function is mainly determined by word order. Both Chinese and

English exhibit topicalization. The entire object NP, for example, can be moved to the beginning

of the sentence in either language. But in Chinese, object topicalization is more restricted when

the subject is quantified: it can happen when the subject is universally quantified, but not when it

is existentially quantified (33-36).

Ambiguous: > and >

(32) Every

Ambiguous: > and >

Ambiguous: > and > 4

In English, subjectverb agreement is not affected by movement; the number and person of the

subject should always agree with the predicate no matter where it occurs. Here, you can assume

that Chinese also follows the subjectverb agreement in the same way that English does.

Figures 1 and 2 show the parse trees of sentences (31) and (33). Topicalization is generally

analysed with gaps. An empty trace is left in the untopicalized position of the object NP, where

4This sentence may seem unambiguously > to some native speakers. But consider this example:
(One book, every student has read, but two books, not necessarily.) The >
reading is in fact available.

Figure 1: English topicalization parse tree: example (31).

Figure 2: Chinese topicalization parse tree: example (33).

every student

Figure 3: Quantifier scope tracking by maintain a list. The parse result of this sentence is > .

S (2) ([,];)

VP ([];)

NP (1) ([];)

every student

S (2) ([]; [,];)

VP ([];)

NP (1) ([];)

every student

Figure 4: The basic idea of quantifier storage.

the gap is introduced. The gapped NP then percolates up the tree, and is finally unified with the

topicalized NP at the left periphery of the sentence.5

Quantifier Storage But if quantifier scoping is a semantic effect, how do we represent it in

syntax? When there is no ambiguity, keeping track of quantifier scope is pretty straightforward.

As shown in figure 3, we can maintain a list-valued feature called a quantifier stack and record

which quantifiers are seen as we ascend whilst building the parse tree. In practice, maintaining this

stack is an instance of a more general process, called beta reduction, that is necessary to manage

semantic expressions in the lambda calculus. We will cover this concept in greater detail in the

tutorials.

To keep track of and resolve scope ambiguities, we can introduce another list: the quantifier

store (represented by ). As shown in figure 4, having this option will allow us to generate parse
trees for multiple readings. At (1), there is an option to store the quantifier in the quantifier store,

and then we can retrieve it at the end (2).

5Although Chinese is an SVO (Subject-Verb-Object) language, there is a means of performing double movement.

A student every book read.

We will ignore these.

(a) (2 marks) Manually convert all readings of the sentences (29) and (30) to logical expressions.

Put your logical forms in section 2(a) of analysis.txt. Use exists and forall for the

quantifiers, and use => and the caret symbol for implication and conjunction.

(b) (10 marks) Implement grammars for the syntax of quantifier scope ambiguity. You dont

need to account for meanings, or for ambiguity in meanings (there should be no syntac-

tic ambiguities). At this point, a correct grammar will produce exactly one parse for every

grammatical sentence. Test your implementation before you move on to the next step.

(c) (10 marks) Augment your grammars to represent meaning and quantifier scope ambiguity.

Marks for question 2(b) will be deducted if your work on this part causes errors in the syn-

tactic predictions. Your grammar should generate more than one parse for each ambiguous

(d) (4 marks) Translate sentences (29) and (30), as you did in the first question.

Operational Instructions

Use the function translate to generate semantic representations of your source sen-

tences. If your sentences can be parsed, translate should open another gralej win-

dow and with all of the translation results.

| ?- translate([a,student,read,every,book]).

You will be prompted as follows to see the next parse.

ANOTHER? y

ANOTHER? y

Answer y to see the next parse until you reach the end. Each time TRALE will open a

new Gralej window. You need to store all of your translation results by repeating the

previous step. A no will be returned when you reach the end of your parses.

Save your translations of sentence (29) as q2d_29_1.grale, q2d_29_2.grale . . . and

your translations of sentence (30) as q2d_30_1.grale, q2d_30_2.grale . . .

Submit a zip file q2d.zip containing all the translation results. You can use this com-

mand: zip -r q2d.zip q2d_*.grale to create the zip file.

Again, dont forget to close all the windows and kill your Gralej processes after you

(e) (4 marks) Again, compare your grammar-based translator with Google Translate. Report

at least one instance of a difference between the translation given by your translator and

Google Translate. Your analysis should be submitted as the section 2(e) in analysis.txt.

CSC 485H/2501H, Fall 2021: Assignment 3

Family name:

Given name:

Student #:

I declare that this assignment, both my paper and electronic submissions, is my own work, and

is in accordance with the University of Toronto Code of Behaviour on Academic Matters and the

Code of Student Conduct.

Signature:

CS: assignmentchef QQ: 1823890830 Email: [email protected]