Self compacting concrete (SCC) is first developed in Japan in the late 1980s in order to overcome the consolidation problems associated with the presence of congested reinforcement. It is also termed as a high performance concrete, as it can flow under its own weight and completely fill the formworks. As the fresh properties of SCC are quite important, mix design of a SCC is performed by considering various workability related fresh properties. Therefore, a well designed SCC should satisfy all requirements of a hardened concrete, besides its superior workability properties. The aim of this research is to assess the effects of some basic ingredients of SCC on the fresh properties of SCC. This will be performed by applying design of experiment techniques and obtaining significant statistical models, which will give valuable information about the effects of the model parameters on the rheology and fresh state characteristics of SCC. In this research program, four different variables / use of fly ash replacement, use of high range water reducing admixture (HRWRA), use of viscosity modifying admixtures (VMA) and water-cementitious material ratio, are considered as the variables of the experimental design. Central Composite Design (CCD), a design of experiment technique, is employed throughout the experimental program and a total of 21 mixtures of concrete are cast. Slump flow, V-funnel, L-box, sieve segregation, initial and final setting time tests are performed, furthermore / to investigate the effects of these variables to the rheology of SCC, relative plastic viscosity and relative yield stress, which are the parameters of Bingham Model are measured with the help of a concrete rheometer. As a result of the experimental program, the fresh state properties of SCC are expressed by mathematical equations. Those equations are then used in order to explain the effects of fly ash replacement, HRWRA and VMA concentration, and the w/cm ratio on the fresh state properties of SCC. According to the derived models, it is stated that the water-cementitious material ratio of the concrete mixture is the most effective parameter on the flowability and passing ability of SCC beside the other parameters utilized in this research as its coefficient was the highest in the related models.

Here we consider penalized regression methods, and extend on the results surrounding the l1 norm penalty. We address a more recent development that generalizes previous methods by penalizing a linear transformation of the coefficients of interest instead of penalizing just the coefficients themselves. We introduce an approximate algorithm to fit this generalization and a fully Bayesian hierarchical model that is a direct analogue of the frequentist version. A number of benefits are derived from the Bayesian persepective; most notably choice of the tuning parameter and natural means to estimate the variation of estimates – a notoriously difficult task for the frequentist formulation. We then introduce Bayesian trend filtering which exemplifies the benefits of our Bayesian version. Bayesian trend filtering is shown to be an empirically strong technique for fitting univariate, nonparametric regression. Through a simulation study, we show that Bayesian trend filtering reduces prediction error and attains more accurate coverage probabilities over the frequentist method. We then apply Bayesian trend filtering to real data sets, where our method is quite competitive against a number of other popular nonparametric methods.

Today’s social and internet networks contain millions or even billions of nodes, and copious amounts of side information (context) such as text, attribute, temporal, image and video data. A thorough analysis of a social network should consider both the graph and the associated side information, yet we also expect the algorithm to execute in a reasonable amount of time on even the largest networks. Towards the goal of rich analysis on societal-scale networks, this thesis provides (1) modeling and algorithmic techniques for incorporating network context into existing network analysis algorithms based on statistical models, and (2) strategies for network data representation, model design, algorithm design and distributed multi-machine programming that, together, ensure scalability to very large networks. The methods presented herein combine the flexibility of statistical models with key ideas and empirical observations from the data mining and social networks communities, and are supported by software libraries for cluster computing based on original distributed systems research. These efforts culminate in a novel mixed-membership triangle motif model that easily scales to large networks with over 100 million nodes on just a few cluster machines, and can be readily extended to accommodate network context using the other techniques presented in this thesis.

published_or_final_version / Social Sciences / Master / Master of Philosophy

Hierarchical models provide a flexible way of modeling complex behavior. However, the complicated interdependencies among the parameters in the hierarchy make training such models difficult. MCMC methods have been widely used for this purpose, but can often only approximate the necessary distributions. Acceptance-rejection sampling allows for perfect simulation from these often unnormalized distributions by drawing from another distribution over the same support. The efficacy of acceptance-rejection sampling is explored through application to a small dataset which has been widely used for evaluating different methods for inference on hierarchical models. A particular algorithm is developed to draw variates from the posterior distribution. The algorithm is both verified and validated, and then finally applied to the given data, with comparisons to the results of different methods.

This paper studies the use of Monte Carlo simulation techniques in the field of econometrics, specifically statistical inference. First, I examine several estimators by deriving properties explicitly and generate their distributions through simulations. Here, simulations are used to illustrate and support the analytical results. Then, I look at test statistics where derivations are costly because of the sensitivity of their critical values to the data generating processes. Simulations here establish significance and necessity for drawing statistical inference. Overall, the paper examines when and how simulations are needed in studying econometric theories.

Collaborative filtering based recommender systems use information about a user's preferences to make personalized predictions about content, such as topics, people, or products, that they might find relevant. As the volume of accessible information and active users on the Internet continues to grow, it becomes increasingly difficult to compute recommendations quickly and accurately over a large dataset. In this study, we will introduce an algorithmic framework built on top of Apache Spark for parallel computation of the neighborhood-based collaborative filtering problem, which allows the algorithm to scale linearly with a growing number of users. We also investigate several different variants of this technique including user and item-based recommendation approaches, correlation and vector-based similarity calculations, and selective down-sampling of user interactions. Finally, we provide an experimental comparison of these techniques on the MovieLens dataset consisting of 10 million movie ratings.

Invasive plants have a significant detrimental effect on ecosystems globally, with impacts estimated at millions of dollars per invasive species each year. Biological control has long been used as a management tool for invasive plants, as it is considered a long–term cost–effective control strategy. Surprisingly, the impact of biological agents is rarely quantified. Any form of impact evaluation is generally conducted soon after agent release and establishment; with few studies examining the impact of the agents on the population dynamics of the invader, particularly once the agents have been established for a long time. The aim of the research in my thesis is to evaluate the biological control program of groundsel bush (Baccharis halimifolia L. Asteraceae) in Australia. The groundsel bush biological control agents were released up to 40 years ago and no quantitative assessment of agent impact has ever been conducted, despite the fact that the program has cost about $9.6 million. More specifically, the overall aim of this thesis is to investigate the impact of the released biological control agents on individual plants and populations of groundsel bush. In addition, my thesis aims to examine the impacts of climate as a potential confounding factor of the biological control program. My thesis provides a unique example of biological control evaluation by using a combination of observational damage studies, insect exclusion experiments, and statistical, population and climate modelling to assess, a posteriori, the effectiveness of biological control. This is the first time a long term biological control program has ever been evaluated. To assess the efficacy of the agents, I conducted a large field survey to examine whether the agents were distributed throughout the entire range of groundsel bush and if any biotic or abiotic factors influenced their effectiveness. In addition to this, I assessed the effect of the agents on the growth, survival and fecundity of individual plants under field conditions, and subsequently population growth rate. To do this, I used statistical models of observed effects of biological control agent damage and insect exclusion experiments on plant growth and fecundity to parameterise matrix population models. My results indicate that the groundsel bush biological control agents may be patchy in their effectiveness due to factors such as rainfall and plant size. At their current rate of damage, the groundsel bush biological control agents do not reduce plant growth or fecundity significantly. However, simulation models demonstrated that the agents have the potential to reduce individual plant and population growth when damage is at high levels. A reduction in an invader’s population growth rate, following the introduction of biological control agents, does not necessarily signify that the agents were responsible for the reduction. Factors such as land clearing, chemical and mechanical control, ecosystem health and climate may reduce populations of invasive plant species. With this in mind, I developed a series of climate models to examine how the favourability for growth of groundsel bush may change under different climate scenarios. The climate simulations demonstrated that the distribution and abundance of groundsel bush populations may have contracted in the past 50 years (post–biological control agent release) due to changing rainfall and temperature patterns. The results of the research in my thesis clearly show the need for thorough biological control evaluations, and for detailed data to be collected on the target plant’s demography and population sizes pre- and post-agent release. At a minimum, this should enable biological control practitioners to determine some level of agent impact and demonstrate support for further agent releases or integrative management strategies if necessary. Groundsel bush is a significant invader in Europe where biological control has not yet been carried out. Lessons from the evaluation of the Australian biological control program could be applied to new biological control programs elsewhere such as Europe. Overall, my research findings contribute to a better understanding of how to best evaluate a post-release biological control program, using groundsel bush as a case study. This is the first study to demonstrate an effective set of strategies and quantitative tools to evaluate a biological control program, which can be similarly applied to any biological control program and contributes significantly to an area of biological control which has only recently received significant attention.

Diss. (sammanfattning) Umeå : Univ., 2008. / Härtill 5 uppsatser.

Thesis--University of Oklahoma. / Bibliography: leaves 79-81.