Tableau d’honneur de la Faculté des études supérieures et postdoctorales, 2016-2017. / La Classification internationale des maladies (CIM) est le principal système de classification utilisé pour les activités de surveillance des blessures, mais ne contient pas d’information directe sur la gravité des blessures. Des mesures de gravité des blessures basées sur la CIM peuvent être dérivées empiriquement ou appariées, mais aucune approche n’a été formellement recommandée, notamment pour les traumatismes craniocérébraux (TCC). Deux études ont été menées afin de combler cette lacune. L’objectif de la première étude consistait à examiner de manière systématique les résultats des études qui ont évalué les performances des mesures de gravité des blessures basées sur la CIM pour prédire la mortalité intra-hospitalière chez les blessés hospitalisés. Cet examen systématique suggère que les mesures de gravité des blessures dérivées empiriquement des codes CIM procurent des capacités de discrimination supérieures à celle des autres candidats examinés. L’ICISS-Multiplicative et le TMPM-ICD offrent une capacité de discrimination plus élevée que les autres mesures considérées. L’objectif de la seconde étude était de comparer la validité prédictive des mesures de gravité des blessures basées sur la CIM pour prédire la mortalité intra-hospitalière et l’admission en unité de soins intensifs chez les patients hospitalisés à la suite d’un TCC au Québec. Nos résultats indiquent que les mesures de gravité des blessures basées sur la CIM affichent une validité acceptable pour prédire la mortalité intra-hospitalière et l’admission en unité de soins intensifs chez les patients hospitalisés à la suite d’un TCC. De manière générale, l’ICISS-Multiplicative a surclassé les autres candidats pour prédire la mortalité à l’hôpital et les admissions en unité de soins intensifs. Sur la base des résultats de ces deux études, nous recommandons que l’ICISS-Multiplicative soit utilisé pour calculer la gravité des TCC dans les bases de données médico-administratives dans lesquelles les diagnostics sont codifiés à l’aide de la CIM. / The International Classification of Diseases (ICD) is the main classification system used for the injury surveillance activities, but contains no direct information on the severity of injuries. ICD-based injury severity measures can be derived empirically or mapped to the Abbreviated Injury Scale, but no approach has been formally recommended, in particular for traumatic brain injury (TBI). Two studies were conducted to fill this gap. The objective of the first study was to systematically review the results of studies that have evaluated the performance of the ICD-based injury severity measures to predict in-hospital mortality among injury-related patients. The results of this systematic review suggest that injury severity measures empirically derived from ICD codes provided superior discriminative performance than other mapped candidates. ICISS-Multiplicative and TMPM-ICD predict mortality more accurately than other ICD-based measures. The objective of the second study was to compare the accuracy of ICD-based injury severity measures for predicting in-hospital mortality and intensive care unit (ICU) admission in TBI patients. Our results indicate that all ICD-based injury severity measures showed acceptable discrimination for predicting in-hospital mortality and ICU admission in TBI patients. ICISS-Multiplicative had generally higher predictive accuracy than alternative ICD-based injury severity measures for predicting in-hospital mortality and ICU admission. Based on the results of these two studies, we recommend that the ICISS-Multiplicative should be used for TBI severity adjustment in large routinely collected administrative data where TBI are coded with ICD.

This thesis discusses the similarities and differences between position classifications in public service and a state institution of higher learning.

Call number: LD2668 .T4 1965 D229 / Master of Science

Call number: LD2668 .T4 1965 H112 / Master of Science

No description available.

No description available.

In the field of Natural Language Processing, supervised machine learning is commonly used to solve classification tasks such as sentiment analysis and text categorization. The classical way of representing the text has been to use the well known Bag-Of-Words representation. However lately low-dimensional dense word vectors have come to dominate the input to state-of-the-art models. While few studies have made a fair comparison of the models' sensibility to the text representation, this thesis tries to fill that gap. We especially seek insight in the impact various unsupervised pre-trained vectors have on the performance. In addition, we take a closer look at the Random Indexing representation and try to optimize it jointly with the classification task. The results show that while low-dimensional pre-trained representations often have computational benefits and have also reported state-of-the-art performance, they do not necessarily outperform the classical representations in all cases.

Ηirbet-Qumran lies about 15km south of Jericho. Between 1947 and 1956 eleven caves were discovered that contained thousands of fragments, mainly of prepared animal skin, representing approximately 900 texts many of which are considered copies. Over 100 of these texts are in Aramaic, though many are short fragments. The provenance of these texts is uncertain, but the fact that copies of some of them are known from beyond Qumran indicates that not all the Aramaic texts found at Qumran are likely to have originated there, though Qumran may have been a site where they were copied. Hitherto, this Aramaic corpus has been referred to as a single entity linguistically, but increasingly it is recognized that heterogeneity of textual features is present. Such heterogeneity may provide clues as to the origins of different texts. The current classification of Qumran texts is in terms of the cave they were found in and the order in which they were found: it does not provide any information about textual relationships between texts. In this thesis I first review the literature to identify suitable quantitative criteria for classification of Aramaic Qumran texts. Second, I determine appropriate statistical methods to classify edited Aramaic Qumran texts according to quantitative textual criteria. Third, I establish ‘proof of principle’ by classifying Hebrew bible books. Fourth, I classify the Aramaic Qumran texts as a corpus without reference to external reference texts. Fifth, I investigate the alignment of this internally-derived classification with external reference texts. Sixth, I perform confirmatory analyses to determine the number of different text-type groups within the corpus. Seventh, I use this classification to allocate previously unclassified texts to one of these text-types. Finally, I examine the textual characteristics of these different text-types and discuss what this can tell scholars about individual texts and the linguistic development of Aramaic during Second Temple Judaism as reflected by Qumran.

Thesis (PhD)--University of Stellenbosch, 2005. / ENGLISH ABSTRACT: The fossiliferous Waaipoort Formation (Witteberg Group, Cape Supergroup), Ganikobis Formation (Dwyka Group, Karoo Supergroup) and Whitehill Formation (Ecca Group, Karoo Supergroup) span a protracted period interrupted by a +-45Ma glacial event. This Late Carboniferous glacial event resulted in a prolonged hiatus in sedimentation as well as significant erosion in many southern continents of Gondwana. This study demonstrates how the glacial episode affected faunal composition and changes in palaeoecology in southern Gondwana across this time period. The Waaipoort shows the highest overall faunal diversity with relatively high palaeoniscoid diversity within Ganikobis and Whitehill Formations. This study is unique in correlating Early Carboniferous - Early Permian fish fauna of South Africa and Namibia specifically, as well as other parts of southern Gondwana, and includes maps of possible fish migration routes. This study attempts to refine the body of disjointed knowledge on the taxonomy and distribution of the fish fauna surrounding this period in southern Gondwana. Three new Waaipoort taxa are recognised, and several reassigned. Earlier work on Late Carboniferous fish from Southern Africa has been updated herein to incorporate recent collections. This has allowed the recognition of up to 15 new palaeonoscoid taxa, some of which are endemic to this region. Several new taxa are recognised within the Whitehill Formation fish fauna and a new lectotype for Palaeoniscum. capensis is presented. Detailed taphonomic and sedimentological studies (with 6 lithologs) of fossil localities visited in this broad-based study verify the correlation, and impart information often lost or ignored in pure palaeontological studies on the palaeoecology or palaeoenvironment of the fauna. Systematic relationships discussed here also serve to strengthen the palaeobiogeographical correlations between the Gondwanan continents. / AFRIKAANSE OPSOMMING: Die fossie1draende Waaipoort Formasie (Witteberg Groep, Kaap Supergroep), Ganikobis Formasie (Dwyka Groep, Karoo Supergroep) en Whitehill Formasie (Ecca Groep, Karoo Supergroep) strek oor 'n uitgebreide tydvak wat 'n periode van vergletsering van ongeveer 45 miljoen jaar insluit. Hierdie Laat Karboon ystyd het sedimentasieprosesse lank onderbreek en beduidende erosie in baie van die suidelike kontinente van Gondwana veroorsaak. Die huidige studie toon hoe die vergletsering die faunistiese samestelling en veranderings in die paleo-ekologie van suidelike Gondwana beinvloed het. In totaal toon die Waaipoort die grootste faunistiese diversiteit, terwyl 'n relatief hoe diversiteit ook onder die Palaeoniscoidei van die Ganikobis en Whitehill Formasies voorkom. Hierdie studie is uniek insoverre dit spesifiek data oor die Vroee Karboon - Vroee Perm visfauna van Suid-Afrika en Namibie, asook ander gebiede van suidelike Gondwana korrelleer en kaarte van moontlike migrasieroetes insluit. Die navorsing poog om die groot hoeveelheid losstaande inligting oor die taksonomie en verspreiding van die visfauna in suidelike Gondwana te verwerk en te verfyn. Drie nuwe taksa is uit die Waaipoort geidentifiseer en 'n verdere aantal is geherklassifiseer. Vroeer navorsing oor die Laat Karboon-visse van Suider-Afrika is bygewerk in die lig van resente versamelings. Dit het die herkenning van tot 15 nuwe taksa onder die Palaeoniscoidei, waarvan sommige endemies aan hierdie streek is, moontlik gemaak. 'n Aantal nuwe taksa word onder die visfauna van die Whitehill Formasie aangeteken, saam met 'n nuwe lektotipe van Palaeoniscum capensis. Gedetailleerde tafonomiese en sedimentologiese analises van die vindplekke wat tydens hierdie studie ondersoek is en wat 6 vertikale profiele insluit, ondersteun hierdie korrelasie. Sodoende stel dit inligting beskikbaar wat dikwels gedurende suiwer paleontologiese studies oor die paleo-ekologie of paleo-omgewing van die fauna verlore sou gaan of geignoreer sou word. Die sistematiese verwantskappe wat hier bespreek word dien ook om die paleobiogeografiese ooreenkomste tussen die Gondwana-kontinente te versterk.

published_or_final_version / Ecology and Biodiversity / Doctoral / Doctor of Philosophy