There has been growing recent interest in the field of active learning for binary classification. This thesis develops a Bayesian approach to active learning which aims to minimise the objective function on which the learner is evaluated, namely the expected misclassification cost. We call this approach the expected cost reduction approach to active learning. In this form of active learning queries are selected by performing a `lookahead' to evaluate the associated expected misclassification cost. \paragraph{} Firstly, we introduce the concept of a \textit{query density} to explicitly model how new data is sampled. An expected cost reduction framework for active learning is then developed which allows the learner to sample data according to arbitrary query densities. The model makes no assumption of independence between queries, instead updating model parameters on the basis of both which observations were made \textsl{and} how they were sampled. This approach is demonstrated on the probabilistic high-low game which is a non-separable extension of the high-low game presented by \cite{Seung_etal1993}. The results indicate that the Bayes expected cost reduction approach performs significantly better than passive learning even when there is considerable overlap between the class distributions, covering $30\%$ of input space. For the probabilistic high-low game however narrow queries appear to consistently outperform wide queries. We therefore conclude the first part of the thesis by investigating whether or not this is always the case, demonstrating examples where sampling broadly is favourable to a single input query. \paragraph{} Secondly, we explore the Bayesian expected cost reduction approach to active learning within the pool-based setting. This is where learning is limited to a finite pool of unlabelled observations from which the learner may select observations to be queried for class-labels. Our implementation of this approach uses Gaussian process classification with the expectation propagation approximation to make the necessary inferences. The implementation is demonstrated on six benchmark data sets and again demonstrates superior performance to passive learning.

Objective. Analyze quality improvement (QI) education across US pharmacy programs. Methods. This was a two stage cross-sectional study that inspected each accredited school website for published QI curriculum or related content, and e-mailed a questionnaire to each school asking about QI curriculum or content. T-test and chi square were used for analysis with an alpha a priori set at .05. Results. Sixty responses (47% response rate) revealed the least-covered QI topics: quality dashboards /sentinel systems (30%); six-sigma or other QI methodologies (45%); safety and quality measures (57%); Medicare Star measures and payment incentives (58%); and how to implement changes to improve quality (60%). More private institutions covered Adverse Drug Events than public institutions and required a dedicated QI class; however, required QI projects were more often reported by public institutions. Conclusion. Despite the need for pharmacists to understand QI, it is not covered well in school curricula.

No description available.

No description available.

Molecular nonlinear optical (NLO) properties are extensively studied using both theory and experiment because of their use in myriad applications. Experimental measurements of the most interesting molecules’ NLO properties are difficult, so experimental data for molecules with desirable NLO properties is scarce. Theoretical tools don’t suffer from the same limitations and can provide significant insights into the physico-chemical phenomena underlying the nonlinear responses, can help in interpreting response behaviour of molecules, and can guide design the materials with desirable response properties. Here, I present my work on developing methods for accurately calculating the NLO properties of molecules using the finite field (FF) approach. The first chapter provides a background for the finite field and electronic structure methods used in this dissertation. Chapter two is a thorough investigation of the finite field method. The limitations of the method are highlighted and the optimal conditions for overcoming its drawbacks and obtaining meaningful and accurate results are described. Chapter three presents the first systematic study of the dependence of optimal field strengths on molecular descriptors. The first protocol for predicting the optimal field for the second hyperpolarizability is presented and successfully tested, and the dependence of the optimal field strength for the first hyperpolarizability on the molecular structure is investigated. Chapter four is an assessment of various DFT functionals in calculating the second hyperpolarizabilities of organic molecules and oligomers. This study shows the limitations of conventional DFT methods and the importance of electron correlation to response properties. In chapter five we present a new method of calculating NLO properties using a rational function model that is shown to be more robust and have lower computational cost than the traditional Taylor expansion. Finally, chapter six includes a summary of the thesis and an overview of future work. / Thesis / Doctor of Philosophy (PhD)

The mechanical properties of materials are usually evaluated by performing a tensile or hardness test on the sample. Tensile tests are often time consuming, destructive and need specially prepared specimens. On the other hand, there is no direct theoretical correlation between the hardness number and the mechanical properties of a material although phenomenological relationships do exist. The advantages of indentation techniques are that they are non-destructive, quick, and can be applied to small material samples and localized in fashion. Mechanical properties are typically determined from spherical indentation load-depth curves. This process is again a time consuming one and not suitable for situations where a quick assessment is required such as in the sheet metal rolling industry. In the present study, a novel method of measuring mechanical properties of the material by multiple spherical indentations is developed. A series of indentations are made on the substrate with a spherical indenter with different loads. The diameter of the indentation is related to the load applied to determine the mechanical properties of the material, namely the yield strength and the work hardening parameters. To determine the diameter of the indentation quickly, a fiber optic sensing technique is developed. An incident light beam from a semiconductor laser is coupled back into an optical fiber upon reflection from the metal surface. By measuring the diffused light power reflected from the metal surface, the diameter of the indentation is measured. The spherical indentation technique is difficult for real time mechanical property measurement of sheet metal in a processing line. Problems arise as the strip is traveling at 2,000 to 4,000 ft/min (10,000 to 20,000 mm/sec) in the processing line. As a first step in developing a process that could be implemented in a real time processing line, a preliminary study has been conducted for the prediction of yield strength by laser shock processing.

The study investigated the impacts of different strategies of providing teacher written corrective feedback on first semester ESL/EFL students’ writing accuracy and writing quality. Four feedback strategies (indirect feedback, direct feedback, indirect feedback followed by direct feedback with explicit corrective comments, and no feedback) were employed in this study. One hundred twenty-one EFL freshman university students were randomly assigned into four feedback groups (IF, DF, IDECC, NF). Students in each group produced two narrative essays. Teacher feedback was provided in two segments for the first essay and students made two revisions based on the feedback. The errors on each stage of students’ writing were marked and counted to be compared among each stage of the writing and between groups. The results of data analysis showed that the mean number of errors in all three treatment groups decreased in each writing stage. All three treatment groups outperformed the no-feedback control group in each stage of writing in terms of grammatical accuracy and writing quality. There was no difference in the mean number of errors among three treatment groups in the first and second revisions. However, the IDECC group, who received indirect feedback followed by direct feedback with explicit corrective comments, outperformed all other groups in the second revision and in the new essay. The results also showed that the mean number of errors of all three treatment groups decreased in the new essay indicating that there was a long-term effect of teacher corrective feedback on the new essay. The results of the study suggest that providing teacher corrective feedback was effective in reducing students’ grammatical errors on their essays. All three treatment groups also gained in writing quality scores in the new essay indicating that, to a certain extent, there was an effect of teacher corrective feedback on writing quality. The findings are discussed in the context of the related literature. Areas of future research are discussed and practical implications are suggested. / Ph. D.