Hra Fantasy Premier League poskytuje miliónom hráčov po celom svete možnosť stať sa na chvíľu manažérom svojho vlastného klubu. Výsledky a bodové ohodnotenie v hre závisia na správnom predvídaní, ako sa budú hráči chovať v skutočných futbalových zápasoch. Ak by pri tomto rozhodovaní pomáhal software na predikciu a analýzu budúcich výkonov hráčov, výsledky v hre sa môžu rapídne zlepšiť. Táto diplomová práca sa zaoberá návrhom a implementáciou predikčného modelu, ktorý využíva neurónové siete na predikcie časových radov počas celej sezóny v hre. Boli použité metódy na spracovanie dát o hráčoch a kluboch za posledné 4 sezóny. Výkonnosť a presnosť predikčných metód boli testované na dátach z poslednej sezóny Premier League a predikcie algoritmu sa vo väčšine prípadov blížili realite. Ak by sa užívateľ držal predikčného modelu v hre stopercentne, získal by väčší počet bodov ako bežný hráč, ktorý žiadny predikčný model nepoužíva.

Bakgrund: Natrium (Na+) är en viktig elektrolyt i kroppen, och analyseras bland annat för att kunna bedöma patientens tillstånd och för att avgöra om akut behandling är nödvändigt. Analysen av Na+ på intensivvårdsavdelningen (IVA) i Örebro sker med hjälp av GEM Premier 5000, vilket är ett patientnära instrument som använder direkt metod för analys av helblod. Vid patientprovsjämförelse skickas provet vidare till centrallaboratoriet där plasman analyseras genom indirekt metod på Advia Chemistry XPT. Avvikelse mellan metoderna får inte överstiga 3%, annars måste orsaken utredas.  Syfte: Syftet med arbetet är att undersöka om det finns en systematisk skillnad på natrium-resultat mellan patientnära instrument, Gem Premier 5000 och centrallaboratoriets instrument, Advia Chemistry XPT hos olika patientgrupper. Metod: Mätning utfördes på blodprover tagna i litium-heparin rör på 60 deltagare, varav 30 var friska blodgivare (grupp 1) och resterande 30 bestod av inneliggande patienter (IVA) samt njurdialys-patienter, (grupp 2). Proverna analyserades för natrium i helblod på GEM Premier 5000 och strax därefter analyserades natrium, albumin, totalt protein, C-reaktivt protein (CRP), glukos och triglycerider i plasma på Advia Chemistry XPT. Resultat: Advia Chemistry XPT gav en högre medelkoncentration av Na+ (139 mmol/L) än GEM Premier 5000 (138 mmol/L) sett till samtliga deltagare. Procentuella skillnaden för natrium mellan metoderna översteg 3% för 3 deltagare i grupp 1 respektive hos hälften av deltagarna i grupp 2.   Slutsats: Na+ resultat på Advia Chemistry XPT var högre än på GEM Premier 5000 för alla deltagare. Skillnaden var större hos patienter med hög grad av sjuklighet. Detta antyder att nuvarande acceptabla avvikelse på 3% bör höjas till 5%, för att antal avvikande värden ska reduceras till nästan samma för båda grupper. Detta måste övervägas och implementeras i verksamheten. / Background: Sodium (Na +) is an important electrolyte in the body, and is analyzed, among other things, to be able to assess the patient's condition in the intensive care unit (IVA) and to determine if emergency treatment is necessary. The analysis of Na + on IVA is done with the help of GEM Premier 5000, which is a patient-centered instrument and uses a direct method for analysis of whole blood. For patient sample comparison, the sample is sent to the central laboratory where the plasma is analyzed by indirect method on Advia Chemistry XPT. Deviation between the methods must not exceed 3%, otherwise the cause must be investigated. Aim: The aim of the study is to investigate whether there is a systematic difference in Sodium results between patient-related instruments, Gem Premier 5000 and the central laboratory's instrument, Advia Chemistry XPT in different patient groups. Method: Measurement was performed on blood samples taken in Lithium Heparin tubes of 60 participants, of which 30 were healthy blood donors (group 1) and the remaining 30 consisted of inpatients (IVA) and kidney dialysis patients, (group 2). The samples were analyzed for sodium on GEM Premier 5000 and shortly thereafter for sodium, albumin, total protein, C-reactive protein (CRP), glucose and triglycerides on Advia Chemistry XPT. Results: Advia Chemistry XPT gave a higher concentration of Na + (139 mmol / L) than GEM Premier 5000 (138 mmol / L) for all participants. The percentage difference of Na between the methods differed for 3 participants in group 1 while it differed for half of the participants in group 2. Conclusion: Na + results on Advia Chemistry XPT were higher than on GEM Premier 5000 for all participants. The difference was greater in patients with a high degree of morbidity. This suggests that the current acceptable deviation of 3% should be increased to 5%, in order to reduce the number of deviating values ​​to almost the same for both groups. This must be taken into account and implemented in the business.

The premise of this report is to delve into sports betting and whether favourable opportunities can be found, more specifically focusing on over and under odds for number of goals scored in football games of the Premier League. Using historical data from football matches several models are developed, the characteristics of goals warranting the use of probability based Poisson and Negative Binomial models, as well as Bayesian Poisson regression for goal predictions. Once these models were developed odds was found and compared to bookmakers, the results indicated that all models, to varying degrees, find favourable opportunities and profitable betting strategies can be identified. This suggests that bookmakers do not always price betting products according to their true probabilities likely due to book balancing and informational asymmetries. Furthermore it indicates that there is a presence of inefficiencies in the sports betting market. / Den här rapporten kommer djupdyka i betting och huruvida gynnsamma möjligheter kan hittas. Mer specifikt kommer ett fokus ligga på över/under odds för antalet mål i fotbollsmatcher i engelska Premier League. Genom att använda historisk data från fotbollsmatcher utvecklas flera olika statistiska modeller för att förutspå antalet mål i fotbollsmatcher. Skattning av Poisson och Negativ Binomial fördelningar samt utvecklandet av en Bayesiansk Poisson regressionsmodell motiveras av egenskaperna hos antalet mål i fotbollsmatcher. Med dessa modeller, beräknas odds för flera framtida matcher inom Premier League och dessa jämfördes med odds som ges av bettingbolag. Resultaten indikerar att alla modeller kan, i olika stor utsträckning, hitta gynnsamma möjligheter och lönsamma betting strategier kan identifieras. Detta tyder på att bettingbolag inte alltid sätter sina odds enbart baserat på den faktiska sannolikheten, vilket troligtvis beror på att bolagen balanserar sina böcker samt informationsasymmetrier. Dessutom indikerar resultatet på att det finns faktorer på bettingmarknaden som gör marknaden ineffektiv.

Le présent mémoire propose une investigation approfondie de l’analyseur Alloy afin de juger son adaptabilité en tant que vérificateur de modèles. Dans un premier temps, l’étude dresse un tableau comparatif de six vérificateurs de modèles, incluant Alloy, afin de déterminer lequel d’entre eux est le plus apte à résoudre les problématiques de sécurité fonctionnelle posées par les systèmes d’information. En conclusion de cette première phase, Alloy émerge comme l’un des analyseurs les plus performants pour vérifier les modèles sur lesquels se fondent les systèmes d’information. Dans un second temps, et sur la base des problématiques rencontrées au cours de cette première phase, l’étude rapporte une série d’idiomes pour, d’une part, présenter une manière optimisée de spécifier des traces et, d’autre part, trouver des recours afin de contourner les limitations imposées par Alloy. À ces fins, le mémoire propose deux nouveaux cas d’espèce, ceux d’une cuisinière intelligente et d’une boîte noire, afin de déterminer si oui ou non l’analyseur est capable de gérer les systèmes dynamiques possédant de nombreuses entités avec autant d’efficacité que les systèmes qui en possèdent moins. En conclusion, le mémoire rapporte que Alloy est un bon outil pour vérifier des systèmes dynamiques mais que sa version récente, DynAlloy, peut être encore mieux adapté pour le faire puisque précisément conçu pour faire face aux spécificités de ce type de système. Le mémoire s’achève sur une présentation sommaire de ce dernier outil.

Thèse numérisée par la Direction des bibliothèques de l'Université de Montréal.

Thèse numérisée par la Division de la gestion de documents et des archives de l'Université de Montréal.

Thèse diffusée initialement dans le cadre d'un projet pilote des Presses de l'Université de Montréal/Centre d'édition numérique UdeM (1997-2008) avec l'autorisation de l'auteur.

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

La manipulation est présente au cœur de l’œuvre d’Albert Camus. Elle s’exerce tant au niveau des personnages qu’au niveau du lecteur. Jean-Baptiste Clamence, dans La Chute, est un personnage extrêmement manipulateur. A la fin de son discours, il énonce ce que furent ses techniques de manipulation. A la lumière de La Chute, on découvre une nouvelle facette de l’œuvre d’Albert Camus, celle de la manipulation.Il existe différents manipulateurs. Il y a ceux qui le font avec un immense plaisir ou ceux qui manipulent pour le bien d’autrui. La manipulation s’exerce parfois au service d’un autre manipulateur. Des événements historiques peuvent être générateurs de manipulation en mettant ces personnages en action.Les personnages manipulés réagissent aussi de différentes manières. On peut dissocier les inconscients des conscients non consentants qui se battent pour s’en libérer. Il existe aussi des personnages conscients de leur manipulation et consentants de l’être.A l’adresse du lecteur, Albert Camus utilise différentes techniques comme le jeu sur les duos, les références à la Bible ou à ses différents écrits. Albert Camus met en place un processus nécessaire de compréhension qui rend le jugement impossible. La Chute apparaît comme le mode d’emploi de la manipulation. A travers les techniques de Jean-Baptiste Clamence, on découvre donc celles d’Albert Camus. Le premier Homme est d’ailleurs la seule œuvre exempte de manipulation et marque un tournant dans l’œuvre d’Albert Camus. / Manipulation is present in Albert Camus’s work. It acts at the level of both the characters and the reader. Jean-Baptiste Clamence, from The Fall, is an extremely manipulative character. At the end of his speech, he lists and describes his manipulation techniques. The Fall reveals a new aspect of Albert Camus’s work : manipulation.There are several kinds of manipulator: those who find personal pleasure in such behaviours, and those who act for the good of others. Manipulation may sometimes benefit another manipulator. Historical events may generate manipulation; they activate Camus’s characters.Manipulated characters also react in different ways. Some are unaware, others may be aware but are non-consenting and fight to free themselves, and other characters are aware of being manipulated and accept it.At the reader’s level, Albert Camus uses different techniques such as the play between duos, references to the Holy Bible or tho his own writings. Albert Camus creates an understanding which makes judgement impossible. The Fall almost resembles an instruction manual for manipulation. Through Jean-Baptiste Clamence’s techniques, we discover those of Camus himself. The First Man is the only one of Camus’s books that is free of manipulation, and in this respect represents a turning-point in his work.

The goal of this study was to improve English language proficiency of undergraduate second-language learners (SLLs) through the use of literacy-based assistive technology (AT). Both current literature and the Universal Design for Learning (UDL) model suggest that literacy-based AT, while traditionally designed to assist students with learning disabilities, can benefit learners studying in a second language. This study adapted the Times Series Concurrent Differential (TSCD) research methodology to test the efficacy of AT for second language learners. TSCD involves the collection of a series of participant performance measurements both with and without the aid of AT. The difference between the two sets of measurements represents the impact of the AT. Fifty-four participants (32 SLL, 22 non-SLL) enrolled in a cross-section of Cape Breton University's Shannon School of Business courses participated. The adapted TSCD model was applied through a series of structured reading exercises that alternated use of AT with traditional reading over a full academic term. The reading assignments were drawn from course material and accounted for a small percentage of the class mark. In non-intervention exercises, participants read and reviewed assignments directly from printed course material. In intervention exercises, participants read and reviewed digital copies of the required material with the aid of PDF Equalizer. A secure Moodle site facilitated digital material access, performance measurement, and data management. A multivariate analysis of covariance (MANCOVA) determined a significant effect (9%) of the use of screen-reading software on academic performance of SLLs and a positive but insignificant effect (3%) of the use of screen-reading assistive technology on academic performance of non-SLLs. In addition, more SLL participants reported that the use of screen-reading software improved their reading (84%), listening (75%), and writing (56%) skills as compared to their non-SLL counterparts (36%, 41%, and 27% respectively). The majority of SLLs also reported that the use of the screen-reader had a positive effect on their academic performance (84%), improved their study skills (84%), and increased their confidence (78%) in their English language skills.