Machine translation offers the challenge of automatically translating a text from one natural language into another. Statistical methods - originating from the field of information theory - have shown to be a major breakthrough in the field of machine translation. Prior to this paradigm, many systems had been developed following a rule-based approach. This denotes a system based on a linguistic description of the languages involved and of how translation occurs in the mind of the (human) translator. Statistical models on the contrary use empirical means and may work with very little linguistic hypothesis on language and translation as performed by humans. This had implications for rule-based translation systems, in terms of software architecture and the nature of the rules, which were manually input and lack any statistical feature. In the view of such diverging paradigms, we can imagine trying to combine both in a hybrid system. In the present work, we start by examining the state-of-the-art of both rule-based and statistical systems. We restrict the rule-based approach to transfer-based systems. We compare rule-based and statistical paradigms in terms of global translation quality and give a qualitative analysis of their respective specific errors. We also introduce initial black-box hybrid models that confirm there is an expected gain in combining the two approaches. Motivated by the qualitative analysis, we focus our study and experiments on lexical phrasal rules. We propose a setup allowing to extract such resources from corpora. Going one step further in the integration of rule-based and statistical approaches, we then examine how to combine the extracted rules with decoding modules that will allow for a corpus-based handling of ambiguity. This then leads to the final delivery of this work: a rule-based system for which we can learn non-deterministic rules from corpora, and whose decoder can be optimised on a tuning set in the same domain.

The researcher’s previous experience suggested the use of computer-based design and fabrication tools might enable new models of practice that yield a greater integration between the 3D art and design disciplines. A critical, contextual review was conducted to assess what kinds of objects are being produced by art and design practitioners; what the significant characteristics of these objects might be; and what technological, theoretical and contextual frameworks support their making. A survey of international practitioners was undertaken to establish how practitioners use these tools and engage with other art and design disciplines. From these a formalised system of analysis was developed to derive evaluative criteria for these objects. The researcher developed a curatorial framework for a public exhibition and symposium that explored the direction that art and design practitioners are taking in relation to computer-based tools. These events allowed the researcher to survey existing works, explore future trends, gather audience and peer response and engage the broader community of interest around the field of enquiry. Interviews were conducted with practitioners whose work was included in this exhibition and project stakeholders to reveal patterns and themes relevant to the theoretical framework of this study. A model of the phases that practitioners go through when they integrate computer-based tools into their practice was derived from an existing technology adoption model. Also, a contemporary version of R. Krauss’s ‘Klein Group’ was developed that considers developments in the field from the use of digital technologies. This was used to model the context within which the researcher’s practice is located. The research identifies a form of ‘technologyled- practice’ and an increased capacity for a ‘transdisciplinary discourse’ at the intersection of disciplinary domains. This study will be of interest to practitioners from across the 3D art and design disciplines that use computerbased tools.

No description available.

This thesis discusses the security of sensor networks. First, an overview of the security architectures of two dominant implementations of sensor networks in the market today is presented: the TinyOS stack and the IEEE 802.15.4 stack. Their similarities and differences are explored and their strength and limitations are discussed. Where applicable, comparisons are made with IEEE 802.11 Wireless LAN to highlight improvements and lessons learned. It is pointed out that in general, IEEE 802.15.4 offers better security, but replay protection is effectively missing in today's implementations and access control is poorly implemented. Consequently, TinyOS is still the better option for devices with severe resource constraints. Finally, as a tool to aid in the security analysis of sensor network, the design and implementation of a TinyOS sniffer is presented and captured frames for a simple sensor network application are analyzed for the purpose of validation.

Repression is a function of many types of states, employed from autocracies to democracies, and anything in between. However, transitional states, those between autocracy and democracy show significantly higher levels of repression than other states. In other studies, research has been done to understand what can be a limitation to repressive activity, and promote democratization. In the case of the post-Soviet state, there have been significant systemic issues that have stalled democratization and allowed it to avoid these limitations. Corruption, consolidation of power into one supermajority party, reliance on electoral manipulation, and passionate development of a national identity all contribute to this problem. As Russia pursues its goals of being a great power, it has exposed itself to globalization and moderating factors. In fact, Russia has accepted western principles on certain rights topics, and implemented positive policies domestically, and supported some human rights legislation at the UN. From this involvement in globalization, Russia may experience the transformative pressures it needs to overcome systemic and structural problems.

While digital coding provides us with numerous benefits, it does have some shortcomings, namely the threshold and the leveling-off effects. These problems do not occur with the use of analog coding, however. Hybrid digital-analog coding systems use both digital and analog coding at the same time to take advantage of the strengths of both coding schemes, while avoiding the problems associated with each. In this work, we propose an HDA system for coding correlates sources. We find the optimal power allocation between the digital and analog parts, and verify the advantages of the suggested system by comparing it to its digital and analog counterparts under different scenarios. / October 2016

Pour conférer au caoutchouc naturel de meilleures propriétés mécaniques, quand il s’agit d’applications industrielles spécifiques telles que le pneumatique, il est mélangé entre autres à des charges renforçantes. L’objectif de cette thèse est double : nous nous sommes premièrement intéressés à l’élaboration d’un nouveau type de charges hybrides, composées d’une partie minérale (silice) et organique (polymère). La stratégie que nous avons développée consiste à synthétiser des nanoparticules hybrides de morphologie contrôlée par polymérisation en émulsion ensemencée et à les agréger a posteriori. Nous avons également réalisé l’étude des propriétés mécaniques d’élastomères chargés avec les charges hybrides obtenues. / It’s necessary to use filler to give better mechanical properties to the natural rubber when it is used in the making of tires. The objective of this thesis is twofold: firstly we have developed a new type of fillers composed of mineral part (silica) and organic part (polymer). The strategy that we developed is to synthesize hybrid nanoparticles of controlled morphology by seeded emulsion polymerization and to aggregate them subsequently. The second objective was to study the mechanical properties of elastomers filled with the obtained hybrid entities.

Master of Science / Department of Agronomy / Scott A. Staggenborg / In Kansas, productivity of grain sorghum [Sorghum bicolor (L.) Moench] is affected by weather conditions at planting and during pollination. Planting date management and selection of hybrid maturity group can help to avoid severe environmental stresses during these sensitive stages. The hypothesis of the study was that late May planting improves grain sorghum yield, growth and development compared with late June planting. The objectives of this research were to investigate the influence of planting dates on growth, development, and yield of different grain sorghum hybrids, and to determine the optimal planting date and hybrid combination for maximum biomass and grains production. Three sorghum hybrids (early, medium , and late maturing) were planted in late May and late June without irrigation in Kansas at Manhattan/Ashland Bottom Research Station, and Hutchinson in 2010; and at Manhattan/North Farm and Hutchinson in 2011. Data on leaf area index, dry matter production, harvest index, yield and yield components were collected. Grain yield and yield components were influenced by planting date depending on environmental conditions. At Manhattan (2010), greater grain yield, number of heads per plant, harvest index, and leaf-area were obtained with late-June planting compared with late May planting, while at Hutchinson (2010) greater yield was obtained with late May planting for all hybrids. The yield component most affected at Hutchinson was the number of kernels panicle-1 and plant density. Late-May planting was favorable for late maturing hybrid (P84G62) in all locations. However, the yield of early maturing hybrid (DKS 28-05) and medium maturing hybrid (DKS 37-07) was less affected by delayed planting. The effects of planting dates on growth, development, and yield of grain sorghum hybrids were found to be variable among hybrid maturity groups and locations.

Coercion and repressive legislation are widely recognised measures employed by hybrid regimes as a way of stifling the media. This thesis illustrates the long shadow cast by these measures by examining the impact of transgressions on self-censorship among Ugandan journalists, and how these are weighed against their notion of professionalism. Self- censorship is experienced as an unwanted, but vital practice that moves in tandem with the level of political tension, being an extraordinary rather than general measure. The study was conducted in the summer of 2016, founded by a Minor Field Study scholarship from the Swedish International Development Cooperation Agency (SIDA).

<p>Various engineering systems have become rapidly automated and intelligent as sensing, communication, and computing technologies have been increasingly advanced. The dynamical behaviors of such systems have also become complicated as they need to meet requirements on performance and safety in various operating conditions. Due to the heterogeneity in its behaviors for different operating modes, it is not appropriate to use a single dynamical model to describe its dynamics, which motivates the development of the stochastic hybrid system (SHS). The SHS is defined as a dynamical system which contains interacting time-evolving continuous state and event-driven discrete state (also called a mode) with uncertainties. Due to its flexibility and effectiveness, the SHS has been widely used for modeling complex engineering systems in many applications such as air traffic control, sensor networks, biological systems, and etc.</p><p>One of the key research areas related to the SHS is the inference or estimation of the states of the SHS, which is also known as the hybrid state estimation. This task is very challenging because both the continuous and discrete states need to be inferred from noisy measurements generated from mixed time-evolving and event-driven behavior of the SHS. This becomes even more difficult when the dynamical behavior or measurement contains nonlinearity, which is the case in many engineering systems that can be modeled as the SHS.</p><p>This research aims to 1) propose a stochastic nonlinear hybrid system model and develop novel nonlinear hybrid state estimation algorithms that can deal with the aforementioned challenges, and 2) apply them to safety-critical applications in air traffic control systems such as aircraft tracking and estimated time of arrival prediction, and unmanned aircraft system traffic management.</p>