Le système de simulation et d'avis précoce d'alerte est un outil puissant pour la surveillance de la qualité de l'eau de la rivière Mudan, une rivière importante dans les régions froides du nord-est de la Chine et qui se jette finalement dans la rivière de l'Amour en Russie. Ainsi la qualité de l'eau dans la rivière Mudan est une préoccupation importante non seulement au niveau local et régional,mais aussi au niveau international. L'objectif de cette étude est de créer un système de simulation et d'avis précoce d'alerte pour que la distribution spatio-temporelle de la qualité de l'eau durant les périodes de couverture glaciaire et d'eaux libres soit simulée et visualisée précisément et afin que l'on puisse appréhender la variation spatiale de polluants sur le cours de rivière. La thèse est structurée en 7 chapitres. Dans le premier chapitre nous décrivons le contexte de l'étude et faisons un état de lieu des recherches actuelles. Dans le chapitre Il, la comparaison des modèles principaux disponibles pour l'évaluation de la qualité de l'eau est réaliser ainsi que le choix du meilleur modèle comme base pour créer le système de modélisation. Dans le chapitre Ill, la construction du modèle,les conditions limites requises et les paramètres pour le modèle ont été vérifiés et étalonnés. Une procédure de simulation distribuée est conçue dans le chapitre IV pour améliorer l'efficacité de la simulation. Le chapitre V concerne la programmation et la réalisation la de simulation distribuée et le chapitre VI les techniques fondamentales pour mettre en œuvre le système. Le chapitre VII est la conclusion. Il y a trois points innovants dans ce travail: un modèle bidimensionnel de dynamique de fluides de l'environnement pour la rivière Mudan, une méthode efficace du calcul distribué et un prototype de système de simulation et d'avis précoce d'alerte qui peuvent largement améliorer la capacité de surveillance et de gestion de la qualité de l'eau de la rivière Mudan ou d'autres rivières similaires. / Simulation and Early Warning System (SEWS) is a powerful tool for river water quality monitoring. Mudan River, an important river in northeastern cold regions of China, can run out of China into Russia. Thus, the water quality of Mudan River is highly concerned not only locally andregionally but also internationally. Objective of this study is to establish an excellent SEWS of water quality so that the spatio-temporal distribution of water quality in both open-water and ice-covered periods can be accurately simulated and visualized to understand the spatial variation of pollutants along the river course. The dissertation is structured into 7 chapters, chapter 1 outlines the background of the study and reviews the current progress. Chapter Il compares the main available models for evaluating river water quality so that a better model can be selected as the basis to establish a modeling system for Mudan River. Chapter Ill establishes the model, the required boundary conditions and parameters for the model were verified and calibrated. Chapter IV, a distributed simulation procedure was designed to increase the simulation efficiency. Chapter V discusses more about the programing and operational issues of the distributed simulation. Chapter VI is about the core techniques to implement the system. Chapter VII is the conclusion of the study to summarize the key points and innovations of the study. The study has the following three points as innovation : a two-dimensional environmental fluid dynamics model for Mudan River, an efficient distributed model computational method and a prototype of SEWS, which can greatly improve the capability of monitoring and management of water quality in Mudan River and other similar rivers.

Submitted by JOSIANE SANTOS DE OLIVEIRA (josianeso) on 2017-06-27T13:30:09Z No. of bitstreams: 1 Paulo Fernando Benetti Marcon_.pdf: 962793 bytes, checksum: 8c45fbcf4084b51d6348450029bf5f28 (MD5) / Made available in DSpace on 2017-06-27T13:30:09Z (GMT). No. of bitstreams: 1 Paulo Fernando Benetti Marcon_.pdf: 962793 bytes, checksum: 8c45fbcf4084b51d6348450029bf5f28 (MD5) Previous issue date: 2017-03-31 / Nenhuma / O uso de recursos tecnológicos para auxiliar nas tarefas de ensino e aprendizagem é uma realidade. A disseminação de ambientes virtuais de aprendizado, como meio de promover a realização de cursos on-line, demonstra franca expansão. Além de tarefas que propiciam a ampliação dos meios de ensino, tais sistemas permitem o registro completo de todas as interações dos alunos no decorrer da realização de disciplinas. Essa gama de informação produzida pode ser utilizada para predição de estudantes em situação de risco enquanto a disciplina ocorre, o que para instituições de ensino pode representar redução nos índices de reprovação e evasão. Entretanto o número elevado de variáveis envolvidas, ainda mais quando várias disciplinas são consideradas, dificulta a construção de modelos computacionais eficientes. Desta forma, este trabalho visa investigar a construção de modelos generalistas – treinados com dados de diversas disciplinas disponíveis – contrapondo a construção de modelos individualizados – treinados individualmente com dados de cada disciplina. Para isto um amplo conjunto de dados educacionais foi extraído, obtido de uma instituição de ensino superior, composto de diferentes cursos, disciplinas e períodos letivos, não sendo utilizadas variáveis que invadissem a privacidade dos estudantes. Uma vez definidas as características e transformações dos dados que contribuíam à identificação de insucesso acadêmico no decorrer da disciplina então foram aplicados algoritmos clássicos de Mineração de Dados seguindo ambas as abordagens, generalista e individualizada, e a cada unidade de conteúdo das disciplinas. Os resultados obtidos demonstram vantagens e desvantagens de ambas as abordagens e que dadas as circunstâncias os modelos individualizados podem ser melhores, obtendo taxas de acerto maiores, e que em outras circunstâncias modelos generalistas apresentam um custo menor para a obtenção e manutenção dos modelos preditivos, mesmo com uma queda nos índices de acerto. / The use of technological resources to assist teaching and learning tasks is a reality. The dissemination of virtual learning environments, as a mean of promoting online courses, shows a clear expansion. In addition to tasks that allow the expansion of teaching resources, such systems allow the complete recording of all the interactions of the students inside the courses. This range of information produced can be used to predict at-risk students while the course is taking place, which for educational institutions may represent a reduction in failure and dropout rates. However, the high number of variables involved, especially when several courses are considered, makes it difficult to construct efficient computational models. In this way, this work aims to investigate the construction of generalist models – trained with data from several available courses – counterposing the construction of individualized models – individually trained with data from each course. In this way, a broad set of educational data was extracted, obtained from a higher education institution, composed of different undergraduate programs, courses and academic periods, not using variables that invaded students' privacy. Once the characteristics and transformations of the data that contributed to the identification of academic insuccess during the course were defined, then classical data mining algorithms were applied following both generalist and individualized approaches and to each content unit of the course. The results obtained demonstrate the advantages and disadvantages of both approaches and that given the circumstances the individualized models may be better, obtaining higher hit rates, and that in other circumstances generalist models present a lower cost for the obtaining and maintenance of the predictive models, even with a drop in hit rates.

Mozambique is one of the countries most affected by natural hazards in the world. Therefore it can benefit greatly from enhancements on its early warning system. Cell broadcast, which is a technology able to send simultaneous alert messages to all mobile phones in a geographical area has gained attention of emergency authorities since various catastrophes in the years 2000’s and increasing diffusion of the mobile network. This research has looked into the disaster risk management routines in Mozambique, interviewing relevant institutions, to identify the currently in use early warning system and analyze the circuit of information from detecting a hazard until the transmission of alert messages to the population. The goal of this research has been to identify how alert messages are sent to the population and, based on currently available infrastructure, analyze the possibilities to use Cell Broadcast to target alerts to all subscribers on specific geographic zones. It has been identified that the country already uses a solution that sends SMS to a list of phone numbers registered in a database. Nonetheless, telecommunication operators in the country are willing to cooperate with emergency authorities to design a solution in which Cell Broadcast can be used to strategically target alerts to subscribers at designated areas of risk. In this way, enabling enhanced accuracy and efficiency of the public alert system in Mozambique, with reduced time between detection and the simultaneously delivery of public alert messages to the entire population or only to people located on relevant geographic zones. Furthermore, the results also allowed to speculate on the viability of automated solutions, which can be used in combination to the enhancements that Cell Broadcast can bring to disaster risk management routines.

Questa tesi di ricerca si occupa del ruolo rivestito dai parlamenti nazionali italiano, britannico e tedesco. Analizza il modo in cui questi parlamenti partecipano al processo decisionale ed implementano il diritto dell'Unione europea dopo il Trattato di Lisbona. La ricerca si concentra su un'analisi comparata delle leggi, delle procedure e consuetudini parlamentari al fine di esaminare il ruolo rivestito dai parlamenti nazionali nel contesto europeo. Il nuovo quadro giuridico previsto dal Trattato di Lisbona promuove la creazione di un sistema parlamentare integrato, basato sulle istituzioni europee e sui parlamenti nazionali cui è attribuito un ruolo più incisivo nel processo decisionale europeo, nella convinzione che un loro maggiore coinvolgimento possa contribuire a garantire un livello più efficace di democrazia nel funzionamento complessivo dell'Unione. I parlamenti nazionali possono contribuire a rendere l'U.E. più o meno efficiente. Essi sono chiamati a svolgere un ruolo rilevante nel processo legislativo europeo, in particolare nella fase di formazione delle politiche e del diritto dell’Unione europea (c.d. fase ascendente) e nel monitoraggio dell'esecuzione del principio di sussidiarietà. Il Trattato di Lisbona introduce norme di partecipazione diretta dei parlamenti nazionali nel processo legislativo europeo, trasformandoli in "guardians of subsidiarity". Il Trattato di Lisbona e i relativi Protocolli riconoscono il ruolo della cooperazione interparlamentare, affidando ai parlamenti nazionali il compito di promuovere e organizzare la sua realizzazione all'interno dell'Unione europea. In questa prospettiva le competenze delle commissioni specializzate in affari europei e della COSAC (Conference of Parliamentary Committees for Union Affairs of Parliaments of the European Union) sono ulteriormente potenziate. / This research thesis deals with the role of national parliaments in Italy, United Kingdom and Germany. It analyses the way in which these Parliaments participate in the European Union and implement the Law of the European Union after the Treaty of Lisbon. The research focuses on a comparative analysis of parliamentary procedures, instruments, and practices in order to examine the respective roles of the European Institutions and the national parliaments within the European framework. The new legal framework laid down the Treaty of Lisbon encourages the creation of an integrated parliamentary system, based on the European Parliament and on the national parliaments which are assigned a more incisive role in the European decision-making process, in the belief that these innovations may contribute to guaranteeing a more effective level of democracy in the overall functioning of the Union. The national parliaments can contribute to making Europe more or less effective. They will be called on to play a more important role in the European law-making process, specifically in the pre-legislative dialogue with European institutions and particularly in the monitoring of the enforcement of the subsidiarity principle in European legislation proposals. The Treaty of Lisbon regulations introduce direct participation of national parliaments in the European law-making process, transforming them into the "guardians of subsidiarity". The Treaty of Lisbon and the related protocols recognise and encourage interparliamentary cooperation, entrusting national parliaments with the task of promoting and organising its achievement within the European Union. In this perspective, the competences of the Conference of Community and European Affairs Committees of Parliaments of the European Union (COSAC) are further enhanced. In this thesis, the reasons for overall inclusion of national parliaments in the European Union activities are analysed. The role of national parliaments in the EU according to the specific provisions of the EU treaties is also discussed and the largest part of the work is devoted to the ex ante subsidiarity principle control mechanism (the Early Warning System), which gives the right for the national parliaments to influence the EU legislative process.

Submitted by Silvana Teresinha Dornelles Studzinski (sstudzinski) on 2017-04-19T15:55:11Z No. of bitstreams: 1 Fábio de Oliveira Dias_.pdf: 3212992 bytes, checksum: 687279fc82a1e707a3ba09c241b5e09b (MD5) / Made available in DSpace on 2017-04-19T15:55:11Z (GMT). No. of bitstreams: 1 Fábio de Oliveira Dias_.pdf: 3212992 bytes, checksum: 687279fc82a1e707a3ba09c241b5e09b (MD5) Previous issue date: 2017-02-17 / Nenhuma / A computação móvel e ubíqua tem propiciado o advento de soluções que permitem o monitoramento em tempo real de sinais provenientes de sensores e o seu processamento por aplicações que podem executar ações de acordo com as condições encontradas. Esta característica possibilita o uso da tecnologia para o monitoramento de condições de saúde de pacientes, denominado de cuidados ubíquos. Em diversas situações, a fim de salvar vidas de pacientes, é necessária a análise de seus sinais vitais de forma a prevenir eventuais colapsos. Este trabalho se insere nestas condições, estando voltado para a antecipação de ações de times de resposta rápida baseado em análise preditiva, propondo o modelo Predictvs. Um Time de Resposta Rápida busca prevenir mortes de pacientes que tenham piora clínica fora de ambientes de Unidades de Tratamento Intensivo em hospitais. De forma diversa dos trabalhos relacionados, que se preocupam apenas com ambientes de tratamento intensivo, o modelo Predictvs busca antecipar ações dos times de resposta rápida, através da análise dos sinais vitais dos pacientes com o uso de escores de alerta precoce e regressão linear. A contribuição científica do modelo é dada em virtude da possibilidade de efetuar a predição em tempo real de possíveis situações de colapso dos pacientes através do monitoramento e análise dos sinais vitais. A avaliação do Predictvs foi efetuada com a utilização de cenários, com a implementação de um protótipo e através de diversas simulações. Análises efetuadas com cerca de 228000 medições provenientes de um dataset público apresentaram bons resultados, onde a precisão da predição para a medição seguinte se mostrou bastante alta, atingindo mais de 99% no caso da frequência cardíaca e 100% na saturação de oxigênio arterial, ultrapassando 95% nos demais sinais vitais. Além disso, o índice de falsos negativos foi consideravelmente baixo, atingindo menos de 1% na frequência cardíaca e na saturação de oxigênio arterial. O índice de falsos positivos também foi baixo, embora não tanto quanto o de falsos negativos. No entanto, predições para três ou mais medições futuras mostram queda na precisão (mesmo demonstrando valores de acerto relativamente expressivos, com diversos sinais fisiológicos acima de 98%) e aumento do número de falsos negativos e, principalmente, de falsos positivos. / The mobile and ubiquitous computing has allowed the emergence of solutions that enable real-time monitoring of signals coming from sensors and processing for applications that can perform actions according to the conditions found. This feature enables the use of this technology for monitoring health conditions of patients, called ubiquitous healthcare. In several situations, in order to save his lives, it is necessary to analyze the vital signs of patients to prevent any collapses. This work is part of these conditions and is aimed at anticipating the actions of rapid response teams based on predictive analysis, proposing the Predictvs model. A Rapid Response Team intends to prevent deaths in patients who have clinical deterioration outside of intensive care units in hospitals environments. Differently of related works, which are concerned only with intensive care environments, the Predictvs model seeks to anticipate the actions of teams of rapid response through the analysis of vital signs of patients with the use of early warning scores and linear regression. The scientific contribution of the presented model is that we could better predict possible collapse situations of patients, through the monitoring and analysis of vital signs. The Predictvs evaluation was performed with the use of scenarios, implementation of a prototype and several simulations. Analyzes performed with about 228,000 measurements from a public dataset showed good results, where the accuracy of the prediction for the next measurement was very high, reaching more than 99% in the case of heart rate and 100% in arterial oxygen saturation, surpassing 95% in other vital signs. In addition, the false negative index was considerably lower, reaching less than 1% in heart rate and arterial oxygen saturation. The rate of false positives was also low, although not so much as that of false negatives. However, predictions for three or more future measurements show a drop in accuracy (even showing relatively expressive set values with several physiological signals above 98%) and an increase in the number of false negatives and, mainly, false positives.

國際研究中雖有針對國家級的銀行脆弱性作分析，卻並未定義或預測台灣系統性危機，本研究在這樣的背景下，決定建構台灣本土的銀行業預警系統，建立銀行危機的領先指標，希望不只順應國際潮流，更能發展適合台灣特殊性的模型。本研究利用貝氏網路模型的特殊性: (1)事後值(2)機率特性，以個體化資料著手，建構一總體性模型。故研究者能確切了解個別銀行財務狀況，對個別銀行發出預警。事後值的特性使研究者能同時考慮多項財務比率。另外，利用機率特性，可幫助研究者了解危機的程度，且能做總體的延伸運用。 本研究發展出兩種方法建構總體模型。第一種為百分比法，以危機銀行佔總銀行個數的比率為基礎；第二種為加權平均法，讓機率值高者有較大權數，機率小者有較小權數去建立一加權平均機率值。 將本研究的推論結果和「台灣金融服務業聯合總會委託計畫-台灣金融危機領先指標之研究」比較，顯示本模型的兩種方法皆與危機之發生有相同趨勢，而考慮危機訊號的設定後，方法二加權平均法顯然具備較佳的預測結果。此外相較總體面衝擊產生的危機，本模型在預測能力上，對來自銀行個體面造成的危機預測明顯較優異。 / International organizations defined and predicted country bank crises events without Taiwan, but they happened in Taiwan in the past twenty years. We construct the early warning system for banking crises in Taiwan and develop the specific model suited to our country. Using Bayesian Model’s specialities: (1) posterior value; (2) probability, we build a systematic model based on microeconomic data. So researcher can understand all financial conditions and predict the financial distresses of individual banks. The concept of posteriority lets researchers can consider a lot of financial ratio at the same time. The characteristic of probability makes researcher to extend the model to macroeconomic. We develop two methods to build systematic model. One is Percentage method which is based on the percentage of financial distress banks to all banks. The other one is weighted average method which used large weight in financial distress bank and small weight in financial sound banks. Comparing our results with the report that Taiwan Financial Services Roundtable issued in 2009, our methods have distress trends which link with crisis directly. But weighted average method has a better predict power than percentage method after considering the signals of distress we specify. Besides, our model has a stronger predictive power in crises from individual effect than crises from macroeconomic shocks.

本財務預警模型研究延續Shumway(2001)年所提出的離散型風險模型(Discrete-time Hazard Model)架構，即Shumway 所稱之多期邏輯斯迴歸模型(Multiperiod logistic regression model) ，來建立銀行財務預警模型。不同於Shumway所提出的Log 基期風險式，研究者根據實際財務危機發生機率圖提出Quadratic 基期風險式。由於離散型風險模型考量與時間相依共變量(Time-dependent covariate)，該模型可以納入隨時間變動的的市場與總體變數，這是單期模型無法達到的。實證結果顯示，不論是否有加入總體與市場變數，Quadratic 基期風險式離散型模型在樣本內檢測表現都比單期模型與Log 基期風險式離散型模型好，研究亦顯示樣本外的預測Quadratic基期風險式在大多數情況都優於Log 基期風險式與單期模型 / This paper continues Shumway(2001) studies on discrete time hazard model, the so called multi-period logistic regression model, to develop a bank failure early warning model . Different from log baseline hazard form proposed by Shumway, author present quadratic baseline hazard form based on the pattern of real default rate. By incorporating time-varying covariates, our model enables us to utilize macroeconomic and market variables, which cannot be incorporated into in a one-period model. We find that our model significantly outperforms the single period logit model and Log baseline hazard model with and without the macroeconomic and market variables at in-sample estimation. The improvement in accuracy comes both from the time-series bank-specific variables and from the time-series macroeconomic variables. Our research also shows that quadratic baseline hazard model outperforms Log baseline hazard model and single period logit model in out-of-sample prediction.

No description available.

No description available.

Transmits decision SS.VII/2 of UNEP Governing Council/Global Ministerial Environment. / UN Job no.: N0230044 E. Material type: Resolutions/decisions (UN). Issued under agenda item 1, agenda document A/CONF.199/PC/1.