Complex networks have been used to represent the fundamental structure of a multitude of complex systems from various fields. In the network representation, the system is reduced to a set of nodes and links that denote the elements of the system and the connections between them respectively. Complex networks are commonly adaptive such that the structure of the network and the states of nodes evolve dynamically in a coupled fashion. Adaptive networks lead to peculiar complex dynamics and network topologies, which can be investigated by moment-closure approximations, a coarse-graining approach that enables the use of the dynamical systems theory. In this thesis, I study several contact processes in adaptive networks that are defined by the transmission of node states. Employing moment-closure approximations, I establish analytical insights into complex phenomena emerging in these systems. I provide a detailed analysis of existing alternative moment-closure approximation schemes and extend them in several directions. Most importantly, I consider developing analytical approaches for models with complex update rules and networks with complex topologies. I discuss four different contact processes in adaptive networks. First, I explore the effect of cyclic dominance in opinion formation. For this, I propose an adaptive network model: the adaptive rock-paper-scissors game. The model displays four different dynamical phases (stationary, oscillatory, consensus, and fragmented) with distinct topological and dynamical properties. I use a simple moment-closure approximation to explain the transitions between these phases. Second, I use the adaptive voter model of opinion formation as a benchmark model to test and compare the performances of major moment-closure approximation schemes in the literature. I provide an in-depth analysis that leads to a heightened understanding of the capabilities of alternative approaches. I demonstrate that, even for the simple adaptive voter model, highly sophisticated approximations can fail due to special dynamic correlations. As a general strategy for targeting such problematic cases, I identify and illustrate the design of new approximation schemes specific to the complex phenomena under investigation. Third, I study the collective motion in mobile animal groups, using the conceptual framework of adaptive networks of opinion formation. I focus on the role of information in consensus decision-making in populations consisting of individuals that have conflicting interests. Employing a moment-closure approximation, I predict that uninformed individuals promote democratic consensus in the population, i.e. the collective decision is made according to plurality. This prediction is confirmed in a fish school experiment, constituting the first example of direct verification for the predictions of adaptive network models. Fourth, I consider a challenging problem for moment-closure approximations: growing adaptive networks with strongly heterogeneous degree distributions. In order to capture the dynamics of such networks, I develop a new approximation scheme, from which analytical results can be obtained by a special coarse-graining procedure. I apply this analytical approach to an epidemics problem, the spreading of a fatal disease on a growing population. I show that, although the degree distribution has a finite variance at any finite infectiousness, the model lacks an epidemic threshold, which is a genuine adaptive network effect. Diseases with very low infectiousness can thus persist and prevail in growing populations.

Antibiotic resistant pathogens, such as methicillin-resistant Staphylococcus aureus (MRSA), and vancomycin-resistant enterococci (VRE), are an increasing burden on healthcare systems. Hospital acquired infections with these organisms leads to higher morbidity and mortality compared with the sensitive strains of the same species and both VRE and MRSA are on the rise worldwide including in Australian hospitals. Emerging community infectious diseases are also having an impact on hospitals. The Severe Acute Respiratory Syndrome virus (SARS Co-V) was noted for its propensity to spread throughout hospitals, and was contained largely through social distancing interventions including hospital isolation. A detailed understanding of the transmission of these and other emerging pathogens is crucial for their containment. The statistical inference and mathematical models used in this thesis aim to improve understanding of pathogen transmission by estimating the transmission rates of contagions and predicting the impact of interventions. Datasets used for these studies come from the Princess Alexandra Hospital in Brisbane, Australia and Shanxi province, mainland China. Epidemiological data on infection outbreaks are challenging to analyse due to the censored nature of infection transmission events. Most datasets record the time on symptom onset, but the transmission time is not observable. There are many ways of managing censored data, in this study we use Bayesian inference, with transmission times incorporated into the augmented dataset as latent variables. Hospital infection surveillance data is often much less detailed that data collected for epidemiological studies, often consisting of serial incidence or prevalence of patient colonisation with a resistant pathogen without individual patient event histories. Despite the lack of detailed data, transmission characteristics can be inferred from such a dataset using structured HiddenMarkovModels (HMMs). Each new transmission in an epidemic increases the infection pressure on those remaining susceptible, hence infection outbreak data are serially dependent. Statistical methods that assume independence of infection events are misleading and prone to over-estimating the impact of infection control interventions. Structured mathematical models that include transmission pressure are essential. Mathematical models can also give insights into the potential impact of interventions. The complex interaction of different infection control strategies, and their likely impact on transmission can be predicted using mathematical models. This dissertation uses modified or novel mathematical models that are specific to the pathogen and dataset being analysed. The first study estimates MRSA transmission in an Intensive Care Unit, using a structured four compartment model, Bayesian inference and a piecewise hazard methods. The model predicts the impact of interventions, such as changes to staff/patient ratios, ward size and decolonisation. A comparison of results of the stochastic and deterministic model is made and reason for differences given. The second study constructs a Hidden Markov Model to describe longitudinal data on weekly VRE prevalence. Transmission is assumed to be either from patient to patient cross-transmission or sporadic (independent of cross-transmission) and parameters for each mode of acquisition are estimated from the data. The third study develops a new model with a compartment representing an environmental reservoir. Parameters for the model are gathered from literature sources and the implications of the environmental reservoir are explored. The fourth study uses a modified Susceptible-Exposed-Infectious-Removed (SEIR) model to analyse data from a SARS outbreak in Shanxi province, China. Infectivity is determined before and after interventions as well as separately for hospitalised and community symptomatic SARS cases. Model diagnostics including sensitivity analysis, model comparison and bootstrapping are implemented.

New Zealand’s First World War studies have traditionally focused on the soldier and battlefield experiences. ‘The Rural Home Front’ breaks with that tradition and focuses on the lives of people and the local communities that the soldiers left behind in the predominantly rural region of Taranaki in New Zealand. ‘The Rural Home Front’ is essentially a study of the impact and effects of the First World War on rural society. By focusing on topics and themes such as ‘war enthusiasm’, the voluntary spirit of fund raising and recruiting, conscription, attempting to maintain normality during wartime, responses to war deaths, the influenza epidemic, the Armistice and the need to remember, this thesis argues that civilians experienced the Great War, too, albeit differently from that of the soldiers serving overseas.

Some Epidemic models with fractional derivatives were proved to be well-defined, well-posed and more accurate [34, 51, 116], compared to models with the conventional derivative. An Ebola epidemic model with non-linear transmission is fully analyzed. The model is expressed with the conventional time derivative with a new parameter included, which happens to be fractional (that derivative is called the 􀀀derivative). We proved that the model is well-de ned and well-posed. Moreover, conditions for boundedness and dissipativity of the trajectories are established. Exploiting the generalized Routh-Hurwitz Criteria, existence and stability analysis of equilibrium points for the Ebola model are performed to show that they are strongly dependent on the non-linear transmission. In particular, conditions for existence and stability of a unique endemic equilibrium to the Ebola system are given. Numerical simulations are provided for particular expressions of the non-linear transmission, with model's parameters taking di erent values. The resulting simulations are in concordance with the usual threshold behavior. The results obtained here may be signi cant for the ght and prevention against Ebola haemorrhagic fever that has so far exterminated hundreds of families and is still a ecting many people in West-Africa and other parts of the world. The full comprehension and handling of the phenomenon of shattering, sometime happening during the process of polymer chain degradation [129, 142], remains unsolved when using the traditional evolution equations describing the degradation. This traditional model has been proved to be very hard to handle as it involves evolution of two intertwined quantities. Moreover, the explicit form of its solution is, in general, impossible to obtain. We explore the possibility of generalizing evolution equation modeling the polymer chain degradation and analyze the model with the conventional time derivative with a new parameter. We consider the general case where the breakup rate depends on the size of the chain breaking up. In the process, the alternative version of Sumudu integral transform is used to provide an explicit form of the general solution representing the evolution of polymer sizes distribution. In particular, we show that this evolution exhibits existence of complex periodic properties due to the presence of cosine and sine functions governing the solutions. Numerical simulations are performed for some particular cases and prove that the system describing the polymer chain degradation contains complex and simple harmonic poles whose e ects are given by these functions or a combination of them. This result may be crucial in the ongoing research to better handle and explain the phenomenon of shattering. Lastly, it has become a conjecture that power series like Mittag-Le er functions and their variants naturally govern solutions to most of generalized fractional evolution models such as kinetic, di usion or relaxation equations. The question is to say whether or not this is always true! Whence, three generalized evolution equations with an additional fractional parameter are solved analytically with conventional techniques. These are processes related to stationary state system, relaxation and di usion. In the analysis, we exploit the Sumudu transform to show that investigation on the stationary state system leads to results of invariability. However, unlike other models, the generalized di usion and relaxation models are proven not to be governed by Mittag-Le er functions or any of their variants, but rather by a parameterized exponential function, new in the literature, more accurate and easier to handle. Graphical representations are performed and also show how that parameter, called ; can be used to control the stationarity of such generalized models. / Mathematical Sciences / Ph. D. (Applied Mathematics)

A substantial number of employees in South Africa may soon be out of work as the result of their HIV-positive status. The dismissal of an infected employee may be motivated by the fact that he is considered to be incompetent or incapable of doing the work for which he was employed. Customers and fellow employees may refuse to deal with an infected employee, with the result that the employee is dismissed for economic reasons. The nature of the undertaking's activities may be such that the presence of an infected employee constitutes a health risk. For the purposes of carrying out his duty to create and maintain safe working conditions, the employer dismisses the employee. The employer may even force the infected employee to resign. Measures should be taken to improve the situation of infected employees. They include educating employers and employees about the transmission of the human immunodeficiency virus. / Mercentile Law / LL. M.

Made available in DSpace on 2014-12-17T14:10:19Z (GMT). No. of bitstreams: 1 FredericoLS_DISSERT.pdf: 1177174 bytes, checksum: efe72b5694aaae13f9be30ff705ec1c9 (MD5) Previous issue date: 2010-10-27 / The pair contact process - PCP is a nonequilibrium stochastic model which, like the basic contact process - CP, exhibits a phase transition to an absorbing state. While the absorbing state CP corresponds to a unique configuration (empty lattice), the PCP process infinitely many. Numerical and theoretical studies, nevertheless, indicate that the PCP belongs to the same universality class as the CP (direct percolation class), but with anomalies in the critical spreading dynamics. An infinite number of absorbing configurations arise in the PCP because all process (creation and annihilation) require a nearest-neighbor pair of particles. The diffusive pair contact process - PCPD) was proposed by Grassberger in 1982. But the interest in the problem follows its rediscovery by the Langevin description. On the basis of numerical results and renormalization group arguments, Carlon, Henkel and Schollw?ck (2001), suggested that certain critical exponents in the PCPD had values similar to those of the party-conserving - PC class. On the other hand, Hinrichsen (2001), reported simulation results inconsistent with the PC class, and proposed that the PCPD belongs to a new universality class. The controversy regarding the universality of the PCPD remains unresolved. In the PCPD, a nearest-neighbor pair of particles is necessary for the process of creation and annihilation, but the particles to diffuse individually. In this work we study the PCPD with diffusion of pair, in which isolated particles cannot move; a nearest-neighbor pair diffuses as a unit. Using quasistationary simulation, we determined with good precision the critical point and critical exponents for three values of the diffusive probability: D=0.5 and D=0.1. For D=0.5: PC=0.89007(3), β/v=0.252(9), z=1.573(1), =1.10(2), m=1.1758(24). For D=0.1: PC=0.9172(1), β/v=0.252(9), z=1.579(11), =1.11(4), m=1.173(4) / O processo de contato por par -PCP ? um modelo estoc?stico de n?o equil?brio que se inspira no processo de contato simples -PC e que exibe uma transi??o de fase para um estado absorvente. Embora que o estado absorvente para o PC corresponda a uma ?nica configura??o (estado vazio), o PCP possui infinitas configura??es. No entanto, estudos num?ricos e te?ricos indicam que o PCP pertence a mesma classe de universalidade do PC (classe da percola??o direcionada), mas apresenta uma anomalia na din?mica de propaga??o. Um n?mero infinito de configura??es de estados absorventes surge no PCP, devido a todos os processos de cria??o e aniquila??o que requererem um par de part?culas de vizinhos mais pr?ximos. O processo de contato por par difusivo - PCPD foi proposto por Grassberger em 1982. Por?m, o interesse neste problema segue com a redescoberta por Howard; T?uber (1997), que questionaram a validade da descri??o de Langevin. Com base nos resultados num?ricos e em grupo de renormaliza??o, Carlon; Henkel ; Schollw?ck, (2001), observaram que alguns expoentes cr?ticos no PCPD apresentam valores similares ao da classe PC. Porem, Hinrichsen (2001), mostrou resultados diferentes do caso PCPD, atrav?s da simula??o, para o caso PC, propondo uma nova classe de universalidade. At? hoje existe uma controv?rsia em rela??o a classe de universalidade do PCPD. No PCPD ? necess?rio um par de part?culas vizinhas para os processos de cria??o e aniquila??o, embora as part?culas difundam individualmente. Neste trabalho, estudamos o PCPDP com difus?o de pares, no qual part?culas isoladas n?o podem difundir. Pares vizinhos difundem juntos. Usando simula??o quase-estacion?ria, determinamos com boa precis?o o ponto cr?tico e os expoentes para dois valores da probabilidade de difus?o: D=0.5, e 0.1. Para D=0.5: PC=0.89007(3), β/v=0.252(9), z=1.573(1), =1.10(2), m=1.1758(24). Para D=0.1: PC=0.9172(1), β/v=0.252(9), z=1.579(11), =1.11(4), m=1.173(4)

Made available in DSpace on 2015-03-03T15:15:25Z (GMT). No. of bitstreams: 1 MarceloBS_DISSERT.pdf: 2228867 bytes, checksum: 46ad012b7ecf9d333c9b9a88bbfb0411 (MD5) Previous issue date: 2010-08-12 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico / The diffusive epidemic process (PED) is a nonequilibrium stochastic model which, exhibits a phase trnasition to an absorbing state. In the model, healthy (A) and sick (B) individuals diffuse on a lattice with diffusion constants DA and DB, respectively. According to a Wilson renormalization calculation, the system presents a first-order phase transition, for the case DA > DB. Several researches performed simulation works for test this is conjecture, but it was not possible to observe this first-order phase transition. The explanation given was that we needed to perform simulation to higher dimensions. In this work had the motivation to investigate the critical behavior of a diffusive epidemic propagation with L?vy interaction(PEDL), in one-dimension. The L?vy distribution has the interaction of diffusion of all sizes taking the one-dimensional system for a higher-dimensional. We try to explain this is controversy that remains unresolved, for the case DA > DB. For this work, we use the Monte Carlo Method with resuscitation. This is method is to add a sick individual in the system when the order parameter (sick density) go to zero. We apply a finite size scalling for estimates the critical point and the exponent critical =, e z, for the case DA > DB / O processo epid?mico difusivo (PED) ? um modelo estoc?stico de n?o equil?brio que se inspira no processo de contato e que exibe uma transi??o de fase para um estado absorvente. No modelo, temos indiv?duos saud?veis (A) e indiv?duos doentes (B) se difundindo numa rede unidimensional com uma difus?o constante DA e DB, respectivamente. De acordo com os c?lculos do grupo de renormaliza??o, o sistema apresentou uma transi??o de fase de primeira ordem, para o caso DA > DB. V?rios pesquisadores realizaram trabalhos de simula??o para testar esta conjectura e n?o conseguiram observar esta transi??o de primeira ordem. A explica??o dada era que precis?vamos realizar simula??o para dimens?es maiores. Por isso, neste trabalho tivemos a motiva??o de investigarmos o comportamento cr?tico de um processo de propaga??o epid?mico difusivo com intera??o de L?vy (PEDL) em uma dimens?o. A distribui??o de L?vy tem intera??o de difus?o de todos os tamanhos levando o sistema unidimensional a um sistema de dimens?es maiores. Com isso, poderemos tentar explicar esta controv?rsia que existe at? hoje, para o caso DA > DB. Para este trabalho utilizamos o M?todo de Monte Carlo com ressuscitamento. Este m?todo consiste em acrescentar um indiv?duo doente no sistema quando o par?metro de ordem (densidade de doente) vai ? zero. Aplicamos a t?cnica de an?lise de escala de tamanho finito para determinarmos com boa precis?o o ponto cr?tico e os expoentes cr?ticos ??/v, v e z, para o caso DA > DB

Il y a de nouveaux défis dans l'administration et dans la conception des systèmes pour optimiser la gestion des ressources des applications basées en nuage Cloud Computing. Certaines applications demandent des performances rigoureuses (par exemple, par rapport aux retards et aux limites de la gigue), tandis que d'autres applications présentent des charges de travail en rafale (volatiles). Cette thèse propose un framework inspiré dans un modèle épidémique (et basé sur des Chaînes de Markov à Temps Continu), qui peut reproduire la volatilité de la charge de travail, à savoir les effets de buzz (quand il y a une augmentation soudaine de la popularité d'un contenu) d'un système de Vidéo à la Demande (VoD). Deux méthodes d'estimation (basés sur des heuristiques et des Chaînes de Markov Monte Carlo - MCMC) ont été également proposées dans ce travail, de façon à ajuster le modèle selon les comportements de la charge de travail. Les paramètres du modèle obtenus à partir des procédures d'étalonnage révèlent des propriétés intéressantes du modèle. Basé sur des simulations numériques, la précision des deux procédures a été analysée, en montrant que les deux présentent des performances raisonnables. Toutefois, la méthode MCMC dépasse la performance de l'approche heuristique. Cette thèse compare également le modèle proposé avec d'autres modèles existants, tout en examinant la qualité de l'ajustement de certaines propriétés statistiques sur des traces réelles de la charge de travail. Finalement, ce travail propose une approche probabiliste de provisionnement des ressources, basée sur le Principe de Grandes Déviations (LDP). LDP caractérise statistiquement les effets de buzz, qui causent de la volatilité extrême de la charge de travail. Cette analyse exploite les informations obtenues en utilisant le LPD du système VoD pour la définition des politiques de gestion des ressources. Ces politiques peuvent être intéressantes pour toutes les acteurs dans le nouveau contexte de l'informatique en nuage. / There are some new challenges in system administration and design to optimize the resource management for a cloud based application. Some applications demand stringent performance requirements (e.g. delay and jitter bounds), while some applications exhibit bursty (volatile) workloads. This thesis proposes an epidemic model inspired (and continuous time Markov Chain based) framework, which can reproduce workload volatility namely the "buzz effects" (when there is a sudden increase of a content popularity) of a Video on Demand (VoD) system. Two estimation procedures (heuristic and a Markov Chain Monte Carlo (MCMC) based approach) have also been proposed in this work to calibrate the model against workload traces. Obtained model parameters from the calibration procedures reveal some interesting property of the model. Based on numerical simulations, precisions of both procedures have been analyzed, which show that both of them perform reasonably. However, the MCMC procedure outperforms the heuristic approach. This thesis also compares the proposed model with other existing models examining the goodness-of-fit of some statistical properties of real workload traces. Finally this work suggests a probabilistic resource provisioning approach based on a Large Deviation Principle (LDP). LDP statistically characterizes the buzz effects that causeextreme workload volatility. This analysis exploits the information obtained using the LDP of the VoD system for defining resource management policies. These policies may be of some interest to all stakeholders in the emerging context of cloud networking.

Made available in DSpace on 2014-07-29T16:17:38Z (GMT). No. of bitstreams: 1 Leandro Carvalho Damacena Neto.pdf: 5860174 bytes, checksum: d2948cd4bdd56c4fd40f454247f3de60 (MD5) Previous issue date: 2011-03-11 / Research on the Spanish flu in Goiás aimed to understand the impacts and meanings which accounted for the population. We analyze its symptoms Spanish flu, as well as highlight the imprecision of medicine to define and characterize it, the multiple symptoms diagnosed and the variety of treatments and therapeutic measures. For this, the research is anchored in the records of the press Goiás, in the context of 1918/1919 were lodged with the population and called Advice to people: that is, they were indications of health authorities to combat the Spanish flu. More than a biological problem, the Spanish flu became a social problem, and as such has been analyzed here, from its social representation - ie, the disease constituted a problem that requires an explanation by the company attacked, it is imperative that has a social and cultural. Historicize diseases is one of the ways to understand a society. / A pesquisa sobre a gripe espanhola em Goiás teve como principal objetivo compreender os impactos e os significados que representou para a população. Buscamos analisar a sintomatologia da doença de gripe espanhola, bem como ressaltar a imprecisão da medicina ao defini-la e caracterizá-la, os múltiplos sintomas diagnosticados e a variedade de tratamentos e medidas terapêuticas. Para tanto, a pesquisa ancorou-se nos registros da imprensa goiana, que, no contexto de 1918/1919, foram dirigidos à população e denominados Conselhos ao povo;ou seja, eram indicações das autoridades sanitárias para o combate da gripe espanhola. Mais que um problema biológico, a gripe espanhola se tornou um problema social, e como tal foi aqui analisada, a partir da sua representação social ou seja, a doença constituiu-se um problema que exige uma explicação pela sociedade atacada; é imperativo que tenha sentido social e cultural. Historicizar as doenças é um dos caminhos para se compreender uma sociedade.

Le continuel développement des outils statistiques permet aujourd’hui la modélisation de nombreux phénomènes, toujours plus complexes. En combinant l’approche offerte par des modèles statistiques spécifiques, dit "multiniveaux", et leurs applications à différentes problématiques médicales, cette thèse s’inscrit à la croisée de divers domaines : celui des statistiques, de par la méthodologie sur laquelle reposent les résultats ; mais aussi, celui de l’économie de la santé et de la santé publique en général, au travers des applications présentées.La première partie de cette thèse s’intéresse aux aspects théoriques, et plus particulièrement à l’évolution des méthodologies, du modèle de régression linéaire simple aux modèles multiniveaux pour des liens non nécessairement linéaires. Le déroulé historique de la modélisation mais également les hypothèses, le principe, la stratégie d'analyse et enfin les limites y sont abordés.La seconde partie s’articule autour de deux applications multiniveaux distinctes. La première concerne les déterminants de l'observance et des interruptions de traitement, chez les personnes infectées par le virus du VIH/Sida suivant un traitement par antirétroviraux, dans le contexte du Cameroun. La seconde, quant à elle, s'intéresse au recours aux soins dentaires chez les personnes âgées de 60 ans et plus et vivant en domicile ordinaire. Ces deux applications sont comparables dans leurs méthodologies puisqu’il s’agit de déterminer les comportements étudiés par des variables individuelles habituellement retenues, mais aussi des variables de "contexte" (caractéristiques de l’offre de soins).La troisième partie est consacrée aux applications sur le panel de médecins généralistes et traite deux études autonomes. La première expose les réticences des médecins, et le rôle de leurs aprioris, face aux Recommandations de Bonnes Pratiques (RBP). Elle nous permet de déboucher sur quelques pistes pour améliorer l’usage des RBP en médecine de ville. La seconde analyse la similarité – ou parfois la dissimilarité – entre le cycle d’activité des médecins généralistes et la saisonnalité des épidémies, dans le but d’identifier les facteurs favorisant l’ajustement des médecins généralistes aux variations saisonnières des besoins des patients. Cette étude pourrait permettre, notamment, d’anticiper et de mieux gérer des situations de crise sanitaire, avec l’appui effectif de la médecine de ville. / The continual development of statistical tools allows the modelling of numerous phenomena, including the complex ones. Using a set of statistical techniques and applications, based on the so-called “multilevel” modelling, this thesis deals with different aspects related to the statistical methodology and applications as per health economics and public health.The first part reconsiders the evolution of methodology, starting from the simple linear regression techniques to the more complex multilevel modelling as applied to both the linear and non-linear relations. It addresses issues related to the historical development, the hypotheses, the strategy of analysis, and the scope of applications. The second part presents two distinct multilevel studies. The first concerns the determinants of observance and interruptions of treatment for persons infected by the HIV/AIDS and treated with antiretroviral in Cameroon. The second one focuses on the use of dental services for the elderly. The two studies are methodologically comparable in that, besides integrating the usually retained individual variables, the analyses examine health seeking behaviours, particularly, the utilisation of health care services, while accounting for contextual determinants such as the characteristics of health supply (clinic, department or region).The third part is dedicated to the applications on GPs' panel and contains two different studies. The first one explains the GPs’ behaviours and the role of their aprioris vis-à-vis Clinical Practice Guidelines (CPG), and suggests solutions to overcome their negative attitudes. The second study analyses the similarity – sometimes the differences – between the cycle of GPs’ activity and the seasonality of epidemics with the aim of better understanding determinants favouring the adjustment of the GP in the seasonal variations of the patients needs. This study helps anticipate and manage situations of sanitary crisis, with the effective support of general practitioners.