Linguistics - Theses, Dissertations, and other Required Graduate Degree Essays

Receive updates for this collection

Quantitative authorship attribution: A history and an evaluation of techniques

Date created: 
2005
Abstract: 

I here present a history of the field of quantitative authorship attribution and an evaluation of its techniques. The basic assumption of quantitative authorship attribution is that the author of a text can be selected from a set of possible authors by comparing the values of textual measurements in that text to their corresponding values in each author's writing sample. Over the centuries, many measurements have been proposed, but never before have the majority of these measurements been tested on the same dataset. Until now investigators of authorship have not known which measurements are the best indicators of authorship. Such information is crucial if our current techniques are to be used effectively and if new more powerhl techniques are to be developed. Based on the results of this study, I propose that the best approach to quantitative authorship attribution involves the analysis of many different types of textual measurements.

Document type: 
Thesis
File(s): 
Department: 
Department of Linguistics - Simon Fraser University
Thesis type: 
Thesis (M.A.)

The ditransitive construction in Korean

Author: 
Date created: 
2008
Abstract: 

This thesis investigates the Korean dative construction that consists of the Goal (Indirect Object: IO), an NP marked with a dative marker, and the Theme (Direct Object: DO), an NP marked with an accusative marker. I propose that [IO-DO] is the underlying order and [DO-IO] is derived through scrambling. Further, I propose that the underlying order [IO-DO] is an instantiation of a prepositional locative structure in which the Goal c-commands the Theme. As supporting arguments, I point out that Korean has another locative structure in the form of a double subject construction. I show that the syntactic relationship in the double subject locative structure is similar to the syntactic relationship in the dative construction.

Document type: 
Thesis
File(s): 
Senior supervisor: 
C
Department: 
Dept. of Linguistics - Simon Fraser University
Thesis type: 
Thesis (M.A.)

Prosodic profiles: Suspects' speech during police interviews

Date created: 
2008
Abstract: 

This dissertation presents a descriptive study of the prosodic characteristics of suspects' speech during investigative interviews with police. During police interviews, investigators direct conversation by asking suspects questions and making assertions which place the suspect in the position of responding. Based on whether the suspect is a first-time or repeat offender, and the type of information suspects produce, responses are categorized and examined for their properties of pause, tempo, and pitch. Response categories explored in this study are affirmative, in which suspects confirm information in the investigators' questions or assertions; negative, in which asserted information is rejected; relevant and irrelevant, in which suspects offer information pertaining or not pertaining to the investigators' questions; and confessions. Certain pausal features—response latency, pause-to-speech ratio—are found to differ across response types in both groups. In general, for example, first time suspects pause more than repeat offenders, both before and during turns, particularly when offering relevant information. Among the temporal features, first time suspects' speech and articulation rates are lower when producing relevant information than repeat offenders' rates. Furthermore, first-timers' irrelevant temporal rates are higher than relevant temporal rates. Pitch characteristics show less distinction across response types than pause and tempo, although first-time suspects' pitch values cluster somewhat more neatly within response types than repeat offenders', whose pitch values vary more widely.

Document type: 
Thesis
File(s): 
Senior supervisor: 
N
Department: 
Dept. of Linguistics - Simon Fraser University
Thesis type: 
Thesis (Ph.D.)

Training Spanish Speakers in the Perception and Production of English Vowels

Date created: 
2003
Abstract: 

This thesis investigates the effects of training native speakers of Spanish in the perception and production of the English pairs of vowels /i/-/I/, /u/-/u/, and /a/-/A/ in a regular ESL classroom setting. Thirty-two adult native Spanish speakers, sixteen in the control group and sixteen in the experimental group, participated in the study. The experimental design included a pretest-posttest procedure in order to compare the subject's performance before and after training. Perception was tested using a minimal pair forced choice task including multiple samples of the three pairs of vowels. Production data was provided by reading 15 sentences and a paragraph, all of which contained the target vowel contrasts. Over a three-week training period, the subjects in the experimental group were given instruction on how to identify and produce the English pairs of vowels 111-/I/, /u/- /u/, and /a/-/A/. Their progress was tested through quizzes at the end of each week. No recordings were used during the lessons. The effect of training on perception was demonstrated through a direct comparison between the scores on the pretest and posttest. Analysis reveals a significant improvement (from 60.1% to 83.3%) in the subjects' performance as an effect of training. A mixed design ANOVA with 1 between factor (group) and 2 within factors (vowel and time) shows a significant group x time interaction (p<.05). The effect of training on production was assessed through a category goodness test and a comprehensibility test. Overall, there was no significant improvement as an effect of training. The study provides evidence that Spanish speaking adults who learn ESL can be taught to perceive certain vowels with more accuracy in a regular classroom environment. Accuracy gained for perception in the training was not transferable to production. The study suggest that exposure to the language might have been an important factor affecting these results in production.

Document type: 
Thesis
File(s): 
Department: 
Theses (Dept. of Linguistics) / Simon Fraser University
Thesis type: 
Thesis (Ph.D.)

Negation scope and phrase structure in Japanese

Date created: 
2004
Document type: 
Thesis
File(s): 
Department: 
Theses (Dept. of Linguistics) / Simon Fraser University
Thesis type: 
Thesis (M.A.)

A lexical semantic study of four-character Sino-Japanese compounds and its application to machine translation

Author: 
Date created: 
2007
Abstract: 

Four-Character Sino-Japanese compounds are a productive word formation process in Japanese. There are many morpho-syntactic analyses on these compounds. However, little has been done on their lexical semantic structure. In this thesis I will provide a syntactically motivated classification system for these compounds, and a lexical semantic analysis of their structure. The lexical semantic analysis is extended to a potential application in Japanese-English Machine Translation. A lexical semantic analysis reveals that for compounds with a deverbal head, there is an argument relation between the constituents if the head’s lexical semantic requirement is fulfilled by the non-head constituent, while the relation is adjunct if it is not fulfilled. The constituents of compounds with a regular noun head are in an attributive relation, and the relation cannot seem to be determined by Lexical Semantics. Compounds with a de-adjectival head require more examples to draw firm conclusions because these compounds are rare.

Document type: 
Thesis
File(s): 
Senior supervisor: 
M
Department: 
Dept. of Linguistics - Simon Fraser University
Thesis type: 
Thesis (M.A.)