Les graphes sont des structures mathématiques capable de modéliser certains systèmes complexes.Une des nombreuses problématiques liée aux graphes concerne la détection de communautés qui vise à trouver une partition en sommet d'un graphe en vue d'en comprendre la structure. A titre d'exemple, en représentant des contratsd'assurances par des noeuds et leurs degrés de similarité par une arête,détecter des groupes de noeuds fortement connectésconduit à détecter des profils similaires, et donc a voir des profils à risques.De nombreux algorithmes ont essayé de répondreà ce problème.Une des méthodes est la propagation de labels qui consiste à ce quechaque noeud puisse recevoir un label par un vote majoritaire de ses voisins.Bien que cette méthode soit simple à mettre en oeuvre,elle présente une grande instabilité due au non déterminisme del'algorithme et peut dans certains cas ne pas détecter de structures communautaires.La première contribution  de cette thèse sera de i) proposerune méthode de stabilisation de la propagation de labelstout en appliquant des barrages artificiels pour limiter les possibles mauvaises propagations.Les réseaux complexes ont également comme caractéristique que certains noeuds puissent appartenir à plusieurs communautés, on parle alors de recouvrements.  C'est en ce sens que la secondecontribution de cette thèse portera sur ii) la créationd'un algorithme auquel seront adjointes des fonctions d'appartenancespour détecter de possibles recouvrements via des noeuds candidats au chevauchement.La taille des graphes est également une notion à considérer dans  la mesure où certains réseaux peuvent contenir plusieursmillions de noeuds et d'arêtes.Nous proposons iii) une version parallèleet distribuée de la détection de communautés en utilisant la propagation de labels par coeur.Une étude comparative sera effectuée pour observerla qualité de partitionnement et de recouvrement desalgorithmes proposés. / Graphs are mathematical structures amounting to a set of nodes (objects or persons) in which some pairs are in linked with edges. Graphs can be  used to model complex systems.One of the main problems in graph theory is the community detection problemwhich aims to find a partition of nodes in the graph to understand its structure.For instance, by representing insurance contracts by nodes and their relationship by edges,detecting groups of nodes highly connected leads to detect similar profiles and to evaluate risk profiles. Several algorithms are used as aresponse to this currently open research field.One of the fastest method is the label propagation.It's a local method, in which each node changes its own label according toits neighbourhood.Unfortunately, this method has two major drawbacks. The first is the instability of the method. Each trialgives rarely the same result.The second is a bad propagation which can lead to huge communities without sense (giant communities problem).The first contribution of the thesis is i)  proposing a stabilisation methodfor the label propagation with artificial dams on edges of some networks in order to limit bad label propagations. Complex networks are also characterized by some nodes which may belong to several communities,we call this a cover.For example, in Protein–protein interaction networks, some proteins may have several functions.Detecting these functions according to their communities could help to cure cancers. The second contribution of this thesis deals with the ii)implementation of an algorithmwith functions to detect potential overlapping nodes .The size of the graphs is also to be considered because some networks  contain several millions of nodes and edges like the Amazon product co-purchasing network.We propose iii) a parallel and a distributed version of the community detection using core label propagation.A study and a comparative analysis of the proposed algorithms will be done based on the quality of the resulted partitions and covers.

Chapter 1: A Population Aging Analysis for Canada Using the National Transfer Accounts Approach: This analysis develops a new data set for Canada using an accounting methodology called National Transfer Accounts (NTA). NTA permits building an accounting system that introduces age into national accounts. NTA is consistent with the conventional national accounts and allows the estimation of lifecycle patterns for labour income and consumption from private and public sources. It also allows the calculation of per capita and aggregate lifecycle deficits (LCD) or surpluses (LCS) in an economy. In this chapter we calculate Canada’s per capita and aggregate LCD for 2006. Using demographic projections for the next five decades, we present the aggregate LCD in Canada for the period 2006 to 2056 assuming a constant per capital LCD during this time horizon. The projection results show that labour income needs to increase rapidly or consumption needs to be cut significantly to compensate for the pressure on the aggregate lifecycle deficit as a result of population aging. Chapter 2: An Overlapping Generations Computable General Equilibrium (OLG-CGE) Model with Age-dependent Rates of Time Preference: This analysis develops a methodology to introduce an age-dependent rate of time preference to calibrate an OLG-CGE model with per capita age profile consumption path derived from the NTA framework of Chapter 1. The results show that the economic impact from an aging population will be significant. The living standard will decline by 15% from the present to 2050 and decline by 20% by the year 2105. To reduce such a decline, the Canadian government may introduce policies that could encourage labour force participation. We analyze the impact of: a) an increase in the general labour force participation rate for age 20 to 64; b) an increase in the labour force participation rate for workers aged 50 to 64; c) late retirement. Our results suggest that maintaining the current standard of living will be extremely difficult after 2026. Nevertheless, this would help reduce the economic pressure from population aging. Chapter 3: 9/11 Security Measures and North American Security Perimeter: A Computable General Equilibrium Analysis: The objective of this analysis is to assess the impact of the North American Security Perimeter (NASP) on both the Canadian and the U.S. economies. The NASP is a change in Canada-U.S. security paradigm that would allow the liberalization of the post 9/11 security measures at the Canada-U.S. border. This study applies a multi-sector and multi-region general equilibrium model together with econometric analysis. After simulating the NASP, Canada’s capital market was found to become more attractive. Under the NASP, the welfare of Canada is estimated to increase by $19 billion or 1.8% of GDP, and that of the U.S. is estimated to increase by $32 billion or 0.3% of GDP.

Volcanic subsidence, caused by partial emptying of magma in the subsurface reservoir has long been observed by spaceborne radar interferometry. Monitoring long-term crustal deformation at the most notable type of volcanic subsidence, caldera, gives us insights of the spatial and hazard-related information of subsurface reservoir. Several subsiding calderas, such as volcanoes on the Galapagos islands have shown a complex ground deformation pattern, which is often composed of a broad deflation signal affecting the entire edifice and a localized subsidence signal focused within the caldera floor. Although numerical or analytical models with multiple reservoirs are proposed as the interpretation, geologically and geophysically evidenced ring structures in the subsurface are often ignored. Therefore, it is still debatable how deep mechanisms relate to the observed deformation patterns near the surface. We aim to understand what kind of activities can lead to the complex deformation. Using two complementary approaches, we study the three-dimensional geometry and kinematics of deflation processes evolving from initial subsidence to later collapse of calderas. Firstly, the analog experiments analyzed by structure-from-motion photogrammetry (SfM) and particle image velocimetry (PIV) helps us to relate the surface deformation to the in-depth structures. Secondly, the numerical modeling using boundary element method (BEM) simulates the characteristic deformation patterns caused by a sill-like source and a ring-fault. Our results show that the volcano-wide broad deflation is primarily caused by the emptying of the deep magma reservoir, whereas the localized deformation on the caldera floor is related to ring-faulting at a shallower depth. The architecture of the ring-fault to a large extent determines the deformation localization on the surface. Since series evidence for ring-faulting at several volcanoes are provided, we highlight that it is vital to include ring-fault activity in numerical or analytical deformation source formulation. Ignoring the process of ring-faulting in models by using multiple point sources for various magma reservoirs will result in erroneous, thus meaningless estimates of depth and volume change of the magmatic reservoir(s).

Background. Virtual Reality is a growing sector that provide the most immersive gaming experiences, especially when the locomotion technique natural walking is used. However, it is always limited by the physical play space available for the user. Introducing Impossible Spaces, also called Overlapping Maps, which when combined with Procedural Content Generation based on the users play space could maximize the experience. Objectives. The aim of this thesis is to implement a potential solution for procedurally generated impossible spaces that are sized according to the VR users playspace. Ultimately testing how the Execution Time is affected when subjecting the implementation to different sized play areas and number of maps to overlap. In addition to this, the thesis will examine the play space setup of various experiencedVR users. Methods. The three core algorithms of the implementation - Grid and Maze generation, as well as Portal placement - are evaluated in terms of execution time. A questionnaire was created for gathering data on VR users and their play space setup. Results. Questionnaire gathered 45 results in total. A majority had access to a play space area of 2-5 square meters. The VR users’ experience affected the size ofthe play space. The execution time for the core algorithms showed promising resultsin terms of execution time. Conclusions. Since most VR users do not have a large play space and the proposed solution performed well, it could be used to enhance the VR experience.

No description available.

No description available.

This study considers software techniques for improving performance on clusters of workstations and approaches for designing message-passing middleware that facilitate scalable, parallel processing. Early binding and overlapping of communication and computation are identified as fundamental approaches for improving parallel performance and scalability on clusters. Currently, cluster computers using the Message-Passing Interface for interprocess communication are the predominant choice for building high-performance computing facilities, which makes the findings of this work relevant to a wide audience from the areas of high-performance computing and parallel processing. The performance-enhancing techniques studied in this work are presently underutilized in practice because of the lack of adequate support by existing message-passing libraries and are also rarely considered by parallel algorithm designers. Furthermore, commonly accepted methods for performance analysis and evaluation of parallel systems omit these techniques and focus primarily on more obvious communication characteristics such as latency and bandwidth. This study provides a theoretical framework for describing early binding and overlapping of communication and computation in models for parallel programming. This framework defines four new performance metrics that facilitate new approaches for performance analysis of parallel systems and algorithms. This dissertation provides experimental data that validate the correctness and accuracy of the performance analysis based on the new framework. The theoretical results of this performance analysis can be used by designers of parallel system and application software for assessing the quality of their implementations and for predicting the effective performance benefits of early binding and overlapping. This work presents MPI/Pro, a new MPI implementation that is specifically optimized for clusters of workstations interconnected with high-speed networks. This MPI implementation emphasizes features such as persistent communication, asynchronous processing, low processor overhead, and independent message progress. These features are identified as critical for delivering maximum performance to applications. The experimental section of this dissertation demonstrates the capability of MPI/Pro to facilitate software techniques that result in significant application performance improvements. Specific demonstrations with Virtual Interface Architecture and TCP/IP over Ethernet are offered.

No description available.

The idea of two elements overlapping to create a third element is a very simple idea, yet one imbued with possibility. It can be as simple as two colors combining to create a new color: yellow + blue = green. This new, third element can stand alone, but it always retains traces of the two original elements. This third element is enriched by the two primary elements, and they, in turn, are enriched by this connection. 1 + 1 = 3 The place where two elements come together can become an integral part of both elements. It can become a central space where ideas meet and intermingle. In such a way, a school of architecture and design centers around its studio. The studio is, for the student, the place where living and learning come together. Here, the practicality of materials meets the theoretical concepts of the classroom. It is often, quite literally, "home away from home" for the student, who spends many hours working on studio projects, gathering with students and faculty, trying to bring design theories into practice. With this idea in mind, a Girls' School of Design is postulated. / Master of Architecture

目前台灣地區地籍圖已全面數值化，電子經偉儀等測量儀器亦不斷改良，土地複丈作業已進入數值化作業之時代。然而現今圖解地籍圖數化區土地複丈外業作業仍有以下二項缺失待檢討改進： 一、複丈外業作業無法於實地一次完成。 二、施測方式是否符合測量規範不易檢核。 為解決以上缺失，本研究針對圖解地籍圖數化區土地複丈外業測量作業流程進行程序管控，利用全測站經緯儀連接筆記型電腦，配合「土地複丈外業作業系統」，並增加外業測量程序管控的功能，將土地複丈外業測量程序與規範納入自動化監控，測量人員需依照電腦程序控制指示逐步完成施測，使土地複丈外業測量、計算、電腦套圖、成果檢核於實地一併完成，如此，不僅使測量作業標準化，確保複丈品質，更可縮減作業流程，提高行政效率。 本研究經以臺北縣中和地政事務所管轄之未辦理地籍圖重測之圖解數化區—南勢角段外南勢角小段與南勢角段頂南勢角小段為實驗區，分別針對土地分割複丈及土地鑑界複丈之外業測量程序以電腦管控，計算坐標後並於實地施以電腦套圖，其成果檢核均能符合地籍測量實施規則之規定，且能於實地一次完成土地複丈之外業作業。 關鍵詞：土地複丈、程序管控、電腦套圖、地籍測量 / Nowadays, because of the inventions and constant refinement of the electronic and computerized measuring tools and instruments, cadastral maps of Taiwan region are all made digitally. However, there are still two major failing flaws to land revision which needed to be modified: 1. Land revision is unable to be completed on site in one outing. 2. Whether the methods of measuring are under government rules and regulations are not readily monitored. In order to correct the failing flaws, this study mainly aims to monitor and control the process of land revision survey with the digitizing operation cadastral maps. To do so, a total station is connected to a notebook computer equipped with the “Land Revision Survey System” software along with procedure control functions. The computer will then monitor and control such activities in an automatic fashion. The surveyors only need to follow the steps given by the computer in order to complete the survey. Land revision of measuring, calculating, map overlapping and result inspecting can be completed all together on site in one outing. This not only standardizes the land revision procedures but also increases its accuracies. Most importantly, it curtails the time spent and increases the administrational efficiency. The regions selected for this research are all under the jurisdiction of Junghe Land Office, Taipei County, which covers the areas of Outer-Nanshijiao, and Top-Nanshijiao subsection, Nanshijiao section. They have yet to be re-surveyed and made into digitizing operation maps. The aim of this research is then to revision these areas by division and combination following the standard controlled procedures. All of the processes follow and comply with the rules and regulations to cadastral survey set by the government. And all of the survey activities are completed on site in one outing. Keywords: Land Revision, Controlled Procedures, Map Overlapping, Cadastral Survey