Discrete phylogeographic models enable the inference of the geographic history of biological organisms along phylogenetic trees. Frequently applied in the context of epidemiological modelling, phylogeographic generalised linear models were developed to allow for the evaluation of multiple predictors of spatial diffusion. The standard phylogeographic generalised linear model formulation, however, assumes that rates of spatial diffusion are a noiseless deterministic function of the set of covariates, admitting no other unobserved sources of variation. Under a variety of simulation scenarios, we demonstrate that the lack of a term modelling stochastic noise results in high false positive rates for predictors of spatial diffusion. We further show that the false positive rate can be controlled by including a random effect term, thus allowing unobserved sources of rate variation. Finally, we apply this random effects model to three recently published datasets and contrast the results of analysing these datasets with those obtained using the standard model. Our study demonstrates the prevalence of false positive results for predictors under the standard phylogeographic model in multiple simulation scenarios and, using empirical data from the literature, highlights the importance of a model accounting for random variation.

The study of individual differences provides insights into how person-specific factors influence decision-making, either before, during or after a decision is made. This dissertation examined a specific individual difference in decision-making: decision-making anxiety. With the adoption of a situation-specific approach, a series of three studies allowed for the conceptual definition of this construct, the development of a measure, and the exploration of its role in the decisionmaking process. Study 1 focused on the development and validation of the Decision-Making Anxiety Inventory. The results demonstrated that the 8-item scale is a useful measure of decision-making anxiety, a superordinate construct, best understood by the interrelations of its three factors of anxiety, worry, and emotionality. Moreover, this study situated decision-making anxiety alongside existing decision-making and personality constructs. In Study 2, the relationships between decision-making anxiety and objective and perceived decision-making competence, and perceived decision quality were examined. This study also included crossvalidation from peers. Findings revealed that anxious decision-makers viewed themselves as poor decision-makers who do not make quality decisions. This perception was not supported by the results from objective measures, nor from peer ratings. In Study 3, the role of decisionmaking anxiety was explored in a specific decision-making context: job search. Data was gathered at two time points, two months apart. This study investigated whether decision-making anxiety led to poorer job choice outcomes, via its relationship with job search behaviours. Results demonstrated that decision-making anxiety was a significant negative predictor of job search effort and intensity, and the focused, exploratory, and haphazard job search strategies. However, decision-making anxiety did not predict the more distal outcomes. Overall, this dissertation highlights that decision-making anxiety is a relevant individual difference in decision-making, which appears to influence individuals’ perceptions about their decisionmaking skills, their experience of decision outcomes, and their decision-related behaviours

Pathogens, such as HIV and influenza, evolve in response to the selective pressures of their host environments accumulating changes in their genomes that offer fitness benefits. This selective pressure is characterised by three properties: (1.) it is episodic, tracking changes in the adaptive immune response and drug therapy, (2.) it is directional in that only particular amino acid substitutions are favoured and (3.) it varies between genomic loci. Most previous models have ignored or inadequately addressed some of these phenomena. This work extends recent approaches to modelling episodic directional selection acting on protein-coding sequences. We use inference techniques within the topic model framework to identify loci evolving under natural selection. A notable example of such techniques are the variational Bayesian methods. We show that our approach performs well in terms of specificity and power, and demonstrate its utility by applying it to some real datasets of HIV sequences.

This study considers an expanded meaning of "uncertainty" as it affects decision-makers. The definition adopted is based on a decision-maker who is uncertain, i.e. aware of the insufficiency of her knowledge for the purpose of rationally determining which option to choose. A taxonomy of uncertainties is developed from this definition. The first stage is a Generalized Decision Model, which expands on a standard decision model often assumed in technical works by allowing uncertainty over components of the model that are assumed to be perfectly known in the standard model. These additional potential "subjects" of uncertainty include the feasibility of options, the authority of the decision-maker to effect a choice, membership of and probability distributions over the set of possible future states of the world, and considerations about how the consequences are to be valued. The taxonomy also describes possible "sources" of uncertainty, dividing them into characteristics of the world (e.g. variability), evidence the decision-maker has (e.g. ambiguity or imprecision), or characteristics of the decision-maker himself. Other important ways in which uncertainties can vary is whether they are hard (irreducible in principle) or soft, whether a decision is unique or repeatable, and the role time has in the decision and in the resolving of the uncertainties. A finding of this work is that many uncertainties in addition to the uncertainty in the standard decision model over the future state of the world can keep a procedure for implementing rational choice from being decisive, thus requiring another (nonrational) process to complete the selection of an option. Other insights: (1) Deciding is only part of being rational, and in many instances is not the most important part. (2) Uncertainty may complicate decision-making, but is by no means always bad for the decision-maker. (3) Rationality is inescapably subjective in any implementation. (4) True "decision under certainty" does not exist. (5) Uncertainties vary sufficiently that no single treatment can be prescribed; it is hoped that this work contributes to a survey of the territory of uncertainty that facilitates Smithson's (1988) "suburbanization" or subdivision into smaller tracts to be developed individually.

Timely and appropriate clinical decisions can be lifesaving, and decision support systems could help facilitate this. However, user adoption of clinical decision support systems (CDSS) and their impact on patient care have been disappointing. Contemporary theories in information systems and several evaluation studies have failed to explain or predict the adoption of CDSS. To find out why I conducted a qualitative inquiry using the constructivist grounded theory method. Guided by the theory of planned behaviour, I designed a functional clinical decision support system called DermML. Then, I used it as a stimulus to elicit responses through semi-structured interviews with doctors, a community to which I also belong. Besides the interview data, I also collected demographic data from the participants and anonymous clickstream data from DermML. I found that the clinical community is diverse, and their knowledge needs are varied yet predictable. Using theoretical sampling, constant comparison and iterative conceptualization, I scaled my findings to a substantive theory that explains the difference in practitioners' knowledge needs and predicts adoption based on CDSS type and use context. Having designed DermML myself, the data provided me with design insights that I have articulated as prescriptive design theory. I posit that GT can generate explanatory and predictive theories and prescriptive design theories to guide action. This study eliminates the boundaries between the developers of CDSS, study participants, future users and knowledge mobilization partners. I hope the rich data I collected and the insights I derived help improve the adoption of CDSS and save lives. / Thesis / Doctor of Philosophy (PhD)

No description available.

No description available.

Yes / Decision tree is a simple but powerful learning technique that is considered as one of the famous learning algorithms that have been successfully used in practice for various classification tasks. They have the advantage of producing a comprehensible classification model with satisfactory accuracy levels in several application domains. In recent years, the volume of data available for learning is dramatically increasing. As a result, many application domains are faced with a large amount of data thereby posing a major bottleneck on the computability of learning techniques. There are different implementations of the decision tree using different techniques. In this paper, we theoretically and experimentally study and compare the computational power of the most common classical top-down decision tree algorithms (C4.5 and CART). This work can serve as part of review work to analyse the computational complexity of the existing decision tree classifier algorithm to gain understanding of the operational steps with the aim of optimizing the learning algorithm for large datasets.

No description available.

A growing body of evidence generated from observational data has demonstrated the potential to influence decision-making and improve patient outcomes. For observational evidence to be actionable, however, it must be generated reliably and in a timely manner. Large distributed observational data networks enable research on diverse patient populations at scale and develop new sound methods to improve reproducibility and robustness of real-world evidence. Nevertheless, the problems of generalizability, portability and scalability persist and compound. As analytical methods only partially address bias, reliable observational research (especially in networks) must address the bias at the design stage (i.e., pre-analysis bias) including the strategies for identifying patients of interest and defining comparators. This thesis synthesizes and enumerates a set of challenges to addressing pre-analysis bias in observational studies and presents mixed-methods approaches and informatics solutions for overcoming a number of those obstacles. We develop frameworks, methods and tools for scalable and reliable phenotyping including data source granularity estimation, comprehensive concept set selection, index date specification, and structured data-based patient review for phenotype evaluation. We cover the research on potential bias in the unexposed comparator definition including systematic background rates estimation and interpretation, and definition and evaluation of the unexposed comparator. We propose that the use of standardized approaches and methods as described in this thesis not only improves reliability but also increases responsiveness of observational evidence. To test this hypothesis, we designed and piloted a Data Consult Service - a service that generates new on-demand evidence at the bedside. We demonstrate that it is feasible to generate reliable evidence to address clinicians’ information needs in a robust and timely fashion and provide our analysis of the current limitations and future steps needed to scale such a service.