Thesis--Universiteit van Amsterdam.

Inaug.-dis.--Jena. / Lebenslauf.

Thesis (Ph. D.)--University of Rhode Island, 2006. / Includes bibliographical references (leaves 175-178).

No description available.

L’existence de noms vieil-anglais au genre variable a été signalée encore à la fin du 19e siècle. Alors que le nombre de noms assignés à plusieurs genres dans les dictionnaires du vieil-anglais est très élevé, il n’existe presque aucune étude approfondie du phénomène. La présente thèse tente de répondre à des questions de ce que l'on désigne comme les noms au genre variable, comment ces noms se distinguent d’autres types de fluctuation dans les documents en vieil-anglais et de ce qui se trouve à l’origine de la variation en genre.Après avoir défini la notion de l’accord et son expression dans les marqueurs, on propose une étude détaillée de toutes les occurrences de trente-six emprunts faits au latin et au vieux-norrois d’une part, et de soixante-dix-huit noms indigènes d’autre part, tous assignés à plusieurs genres dans les dictionnaires. Constatant des évolutions parfois importantes dans l’emploi des marqueurs en discours, on cherche à déterminer s’ils interviennent dans la décision des lexicographes ou s’ils reflètent une variabilité interne aux noms étudiés. La variation se présente ainsi sous deux angles et s’explique à travers deux notions, celle du désaccord, où tout lien entre le nom-contrôleur et les cibles d’accord est rompu, et celle de la variabilité, qui soutient le lien d’accord sur le plan cognitif et permet la transition d’un genre à l’autre selon le choix de l’énonciateur. / The existence of nouns of variable genre in Old English was brought to the linguistic community’s attention at the end of the 19th century. Despite the rather high number of nouns assigned to several genders in dictionaries dealing with Old English, to date there has been no substantial study of the phenomenon. This thesis is a usage-based study that explores the notion of nouns of variable gender by distinguishing them from other types of gender-variation in Old English texts. It also explores in detail the factors behind various grammatical gender assignments.It departs from the notion of agreement and its expression in gender-sensitive markers. A corpus of a hundred and fourteen nouns assigned to several genders in the dictionaries – seventy-eight native and thirty-six borrowed from Latin and Old Norse – are analysed in context in order to identify the various factors that influence the lexicographers’ decision-making. Some important developments in the usage of the formerly gender-sensitive markers in the discourse might have influenced the latter to a certain degree. Or, they might be a reflection of variability as an internal feature of the nouns analysed. This study deals with the notion of variation as a cover-term for disagreement, which reflects the discontinuity of the link between a controller and its agreement targets, and, on the other hand, for variability, maintaining the agreement on the cognitive level and allowing the transition from one gender to the other depending on the choice of the speaker.

This study investigates some aspects of grammatical variation in vernacular Jordanian Arabic (JA), namely word order variation and pro(noun)-drop variation. Much previous research on word order and subject expression in Arabic has been hampered by the use of eclectic methodologies (Bakir 1980; Eid 1983; El-Yasin 1985; Fassi Fehri 1993; Aoun & Li 1993; Brustad 2000). Conspicuously rare in contemporary studies of syntactic variation in Arabic are systematic analyses of spontaneous speech data (Edwards 2010: 94; but see e.g., Owens, Dodsworth & Rockwood 2009; Owens, Dodsworth & Kohn 2013). The dearth of quantitative studies of word order variation, as well as pro-drop variation, in colloquial Arabic provides the primary motivation for the present investigation. Drawing on the framework of variationist sociolinguistics (Labov 1972), I conduct an accountable analysis of word order variation, as well as pro-drop variation in a corpus of vernacular Jordanian Arabic recorded in the Irbid metropolitan area in 2014. The corpus is based on over 30 hours of digitized recordings obtained from 30 speakers stratified by age, sex, education, as well as urban/rural origin. I exploit these spontaneous speech data to: (i) assess the frequency of different word order and pro-drop variants in vernacular JA; (ii) ascertain which social and linguistic factors constrain the selection of major word order and pro-drop variants; and (iii) determine whether the apparent time component incorporated into the research design reveals any evidence of change in progress. Distributional and multivariate analyses of 4500 tokens (2049 for word order and 2422 for pro-drop) coded for the aforementioned social factors, in addition to an array of linguistic factors hypothesized to constrain variant choice (e.g., morphloexical class of subject, grammatical person and number, type of clause and transitivity) confirm that word order variation, as well as pro-drop variation, are subject to multiple constraints (Holes 1995; Owens et al. 2013). A first important finding concerns the quantitative preponderance of SV(O) word order in vernacular JA, which competes with less frequent VS(O). Another important finding is that null subject pronouns are the norm in vernacular JA. Statistical analyses of the linguistic factors conditioning the observed variability reveal that transitivity and definite subject pronouns are key predictors of SV(O) word order choice, while switch reference and person and number of subject are key predictors of overt subject pronouns, as determined by the relative magnitude of these effects. Particularly compelling is the social embedding of the variation in the case of word order variation. Age- and sex-differentiations in the data (Labov 1990), in addition to urban-rural split, reveal statistically significant differences, offering provisional indications that alternation between SV(O) and VS(O) word orders is implicated in ongoing change. Younger speakers, women and urban-origin speakers lead in the use of SV(O). The results foreground the utility of empirically accountable analyses of spontaneous speech in elucidating key issues relating to syntactic variation in modern varieties of spoken Arabic. The results generated by this approach reveal new findings not previously available from the intuited, elicited or written material on which much previous work on Arabic has been based.

Le romani, ou romanes, est une langue parlée historiquement par des communautés qui s’autodénomment « Roms », « Sintos » et « Calos ». Dans cette thèse, on se propose d'analyser une page du réseau social Facebook, comportant de nombreux messages en langue romani. Dans un premier temps, nous identifierons les dialectes présents sur la page et étudierons la diversité des pratiques orthographiques. Puis, on analysera comment la variation est prise en compte par les locuteurs, en évoquant les problématiques d'intercompréhension, d'identité et de revitalisation. Le but de notre recherche est de faire un état des lieux des besoins, réussites et représentations des locuteurs du romani afin d’accompagner ce qui pourrait être un processus d’autogestion langagière. / Romani, also known as Romanes, is a language historically spoken by communities who call themselves “Roma”, “Sinti” and “Kalo”. The purpose of this thesis is to study a page of the social network Facebook which includes many messages written in Romani language. First of all, we will identify the dialects used on the page and we will describe the diversity of spelling practices. Then, we wil analyze the way variation is taken into account by the speakers, discussing the issues of mutual understanding, identity and revitalization. Our research focuses on determining the needs, successes and representations of the Romani speakers in order to support a possible linguistic self-management.

Les décès du cancer de la peau sont majoritairement des mélanomes malins. Il est considéré comme l’un des plus dangereux cancer. A ses débuts, les mélanomes malins sont traités avec des simples biopsies et sont complètement curable. Pour cela, une détection précoce est la meilleure solution pour réduire ses conséquences désastreuses. Imagerie médicale telle que la dermoscopie et les caméras à images standard sont les outils disponibles les plus adaptées pour diagnostiquer précocement les mélanomes. Le diagnostic assisté par ordinateur (CAD) est développé dans le but d’accompagner les radiologistes dans la détection et le diagnostic. Cependant, il y a un besoin d’améliorer la précision de la segmentation et de détection des lésions. Dans ce travail, le modèle de Chan et Vese a été adapté pour segmenter davantage les variations à l’intérieur des lésions avec un résultats très encouragent. La deuxième tâche consiste à extraire des caractéristiques afin de discriminer les mélanomes. Deux méthodes ont été développée, une se basant sur l’irrégularité des bords des lésions et l’autre par la fusion des caractéristiques texturales et structurelles. Les résultats ont montrés de bonnes performances avec une précision de 86.54% et de 86.07%, respectivement. / The vast majority of skin cancer deaths are due to malignant melanoma. It is considered as one of the most dangerous cancers. In its early stages, malignant melanoma is completely curable with a simple biopsy. Therefore, an early detection is the best solution to improve skin cancer prognostic. Medical imaging such as dermoscopy and standard camera images are the most suitable tools available to diagnose melanoma at early stages. To help radiologists in the diagnosis of melanoma cases, there is a strong need to develop computer aided diagnosis (CAD) systems. The accurate segmentation and classification of pigment skin lesions still remains a challenging task due to the various colors and structures developed randomly inside the lesions. The current work focused on two main tasks. Inthe first task, a new approach of the segmentation of skin lesions based on Chan and Vese model is developed. The model is adapted to segment the variations of the pigment inside the lesion and not only the main border. The subjective evaluation, applied on a database of standard camera images, obtained a very encouraging results with 97.62% of true detection rate. In the second main task, two feature extraction methods were developed for the analysis of standard camera and dermoscopy images. The method developed for the standard camera skin cancer images is based on border irregularities, introducing two new concepts, which are valleys and crevasses as first and second level of the border irregularity. The method has been implemented on DermIs and DermQues, two databases of standard camera images, and achieved an accuracy of 86.54% with a sensitivity of 80% and a specificity of 95.45%. The second method consisted of a fusion of structural and textural features. The structural features were extracted from wavelet and curvelet coefficients, while the textural features were obtained from the local binary pattern operator. The method has been implemented on the PH2 database for dermoscopy images with 1000-random sampling cross validation. The obtained results achieved an accuracy, a sensitivity and a specificity of 86:07%, 78.93% and 93.25%. Compared to the existing methods, the proposed methods in this work show very good performances.

Mutations affecting the mitochondrial electron transport chain cause numerous neurodegenerative disorders in humans and affect longevity in other organisms. A natural model system to study the relationship between mitochondrial function and aging within an evolutionary or population genetic context has been lacking. Natural populations of Caenorhabditis briggsae nematodes were recently found to harbor mitochondrial genetic variation with likely functional consequences for aging. Specifically, C. briggsae isolates containing high frequencies of a deletion mutation affecting the mitochondrial NADH dehydrogenase 5 (ND5) gene were found to have reduced reproductive fitness and lifespan and elevated levels of mutagenic superoxide. Here, rates of growth and aging and aerobic respiratory capacity were evaluated in several isolates spanning the range of mitochondrial genetic variation in this species. There is considerable variation among isolates for all measured traits, although the observed relationships between isolate-specific trait means and ND5 deletion frequency did not always conform to my expectations. In an effort to determine whether the among-isolate phenotypic variation is due to mitochondrial rather than to nuclear genetic variation, inter-population hybrids of C. briggsae were created and compared to the progenitor isolates. Surprisingly, evidence for paternal mitochondrial inheritance was detected in many of these hybrid lines. Where mitochondrial genomes were maternally inherited as expected, intergenomic epistasis appears to contribute to fitness, longevity, and aging in this species.

Cementitious materials obtained from different sources, while evaluated and classified by the same methods and criteria, often produce concrete with compressive strength variance despite other inputs remaining constant. The focus of this thesis was to enumerate and illustrate the possible compressive strength variation when cementitious material sources are interchanged, and investigate the influence that aggregate can have on this variation. This was accomplished by compiling and analyzing compressive strength data from previous research initiatives, and concluded that coefficient of variation (COV) and range values at the 14-, 28-, and 56-day timeframes due to a cement source change varied between 15.3% and 18.1% and 1,988 psi and 2,728 psi in concrete, and 16.1% and 22.9% and 3,406 psi and 5,884 psi in paste or mortar. Concrete that included supplementary cementitious material (SCM) displayed up to 4.1% higher COV values versus non-SCM mixtures, and specific aggregate/cementitious material combinations influenced compressive strength variability.