Hierarchical models for 2D presence/absence data having ambiguous zeroes: With a biogeographical case study on dingo behaviour

This dissertation is primarily an applied statistical modelling investigation, motivated by a case study comprising real data and real questions. Theoretical questions on modelling and computation of normalization constants arose from pursuit of these data analytic questions. The essence of the thesis can be described as follows. Consider binary data observed on a two-dimensional lattice. A common problem with such data is the ambiguity of zeroes recorded. These may represent zero response given some threshold (presence) or that the threshold has not been triggered (absence). Suppose that the researcher wishes to estimate the effects of covariates on the binary responses, whilst taking into account underlying spatial variation, which is itself of some interest. This situation arises in many contexts and the dingo, cypress and toad case studies described in the motivation chapter are examples of this. Two main approaches to modelling and inference are investigated in this thesis. The first is frequentist and based on generalized linear models, with spatial variation modelled by using a block structure or by smoothing the residuals spatially. The EM algorithm can be used to obtain point estimates, coupled with bootstrapping or asymptotic MLE estimates for standard errors. The second approach is Bayesian and based on a three- or four-tier hierarchical model, comprising a logistic regression with covariates for the data layer, a binary Markov Random field (MRF) for the underlying spatial process, and suitable priors for parameters in these main models. The three-parameter autologistic model is a particular MRF of interest. Markov chain Monte Carlo (MCMC) methods comprising hybrid Metropolis/Gibbs samplers is suitable for computation in this situation. Model performance can be gauged by MCMC diagnostics. Model choice can be assessed by incorporating another tier in the modelling hierarchy. This requires evaluation of a normalization constant, a notoriously difficult problem. Difficulty with estimating the normalization constant for the MRF can be overcome by using a path integral approach, although this is a highly computationally intensive method. Different methods of estimating ratios of normalization constants (N Cs) are investigated, including importance sampling Monte Carlo (ISMC), dependent Monte Carlo based on MCMC simulations (MCMC), and reverse logistic regression (RLR). I develop an idea present though not fully developed in the literature, and propose the Integrated mean canonical statistic (IMCS) method for estimating log NC ratios for binary MRFs. The IMCS method falls within the framework of the newly identified path sampling methods of Gelman & Meng (1998) and outperforms ISMC, MCMC and RLR. It also does not rely on simplifying assumptions, such as ignoring spatio-temporal dependence in the process. A thorough investigation is made of the application of IMCS to the three-parameter Autologistic model. This work introduces background computations required for the full implementation of the four-tier model in Chapter 7. Two different extensions of the three-tier model to a four-tier version are investigated. The first extension incorporates temporal dependence in the underlying spatio-temporal process. The second extensions allows the successes and failures in the data layer to depend on time. The MCMC computational method is extended to incorporate the extra layer. A major contribution of the thesis is the development of a fully Bayesian approach to inference for these hierarchical models for the first time. Note: The author of this thesis has agreed to make it open access but invites people downloading the thesis to send her an email via the 'Contact Author' function.

The differential impact of holocaust trauma across three generations

In the current thesis, the reasons for the differential impact of Holocaust trauma on Holocaust survivors, and the differential intergenerational transmission of this trauma to survivors’ children and grandchildren were explored. A model specifically related to Holocaust trauma and its transmission was developed based on trauma, family systems and attachment theories as well as theoretical and anecdotal conjecture in the Holocaust literature. The Model of the Differential Impact of Holocaust Trauma across Three Generations was tested firstly by extensive meta-analyses of the literature pertaining to the psychological health of Holocaust survivors and their descendants and secondly via analysis of empirical study data. The meta-analyses reported in this thesis represent the first conducted with research pertaining to Holocaust survivors and grandchildren of Holocaust survivors. The meta-analysis of research conducted with children of survivors is the first to include both published and unpublished research. Meta-analytic techniques such as meta-regression and sub-set meta-analyses provided new information regarding the influence of a number of unmeasured demographic variables on the psychological health of Holocaust survivors and descendants. Based on the results of the meta-analyses it was concluded that Holocaust survivors and their children and grandchildren suffer from a statistically significantly higher level or greater severity of psychological symptoms than the general population. However it was also concluded that there is statistically significant variation in psychological health within the Holocaust survivor and descendant populations. Demographic variables which may explain a substantial amount of this variation have been largely under-assessed in the literature and so an empirical study was needed to clarify the role of demographics in determining survivor and descendant mental health. A total of 124 participants took part in the empirical study conducted for this thesis with 27 Holocaust survivors, 69 children of survivors and 28 grandchildren of survivors. A worldwide recruitment process was used to obtain these participants. Among the demographic variables assessed in the empirical study, aspects of the survivors’ Holocaust trauma (namely the exact nature of their Holocaust experiences, the extent of family bereavement and their country of origin) were found to be particularly potent predictors of not only their own psychological health but continue to be strongly influential in determining the psychological health of their descendants. Further highlighting the continuing influence of the Holocaust was the finding that number of Holocaust affected ancestors was the strongest demographic predictor of grandchild of survivor psychological health. Apart from demographic variables, the current thesis considered family environment dimensions which have been hypothesised to play a role in the transmission of the traumatic impact of the Holocaust from survivors to their descendants. Within the empirical study, parent-child attachment was found to be a key determinant in the transmission of Holocaust trauma from survivors to their children and insecure parent-child attachment continues to reverberate through the generations. In addition, survivors’ communication about the Holocaust and their Holocaust experiences to their children was found to be more influential than general communication within the family. Ten case studies (derived from the empirical study data set) are also provided; five Holocaust survivors, three children of survivors and two grandchildren of survivors. These cases add further to the picture of heterogeneity of the survivor and descendant populations in both experiences and adaptations. It is concluded that the legacy of the Holocaust continues to leave its mark on both its direct survivors and their descendants. Even two generations removed, the direct and indirect effects of the Holocaust have yet to be completely nullified. Research with Holocaust survivor families serves to highlight the differential impacts of state-based trauma and the ways in which its effects continue to be felt for generations. The revised and empirically tested Model of the Differential Impact of Holocaust Trauma across Three Generations presented at the conclusion of this thesis represents a further clarification of existing trauma theories as well as the first attempt at determining the relative importance of both cognitive, interpersonal/interfamilial interaction processes and demographic variables in post-trauma psychological health and transmission of traumatic impact.

On the data consumption benefits of accepting increased uncertainty

In the context of learning paradigms of identification in the limit, we address the question: why is uncertainty sometimes desirable? We use mind change bounds on the output hypotheses as a measure of uncertainty, and interpret ‘desirable’ as reduction in data memorization, also defined in terms of mind change bounds. The resulting model is closely related to iterative learning with bounded mind change complexity, but the dual use of mind change bounds — for hypotheses and for data — is a key distinctive feature of our approach. We show that situations exists where the more mind changes the learner is willing to accept, the lesser the amount of data it needs to remember in order to converge to the correct hypothesis. We also investigate relationships between our model and learning from good examples, set-driven, monotonic and strong-monotonic learners, as well as class-comprising versus class-preserving learnability.

Acoustic keyword spotting in speech with applications to data mining

Keyword Spotting is the task of detecting keywords of interest within continu- ous speech. The applications of this technology range from call centre dialogue systems to covert speech surveillance devices. Keyword spotting is particularly well suited to data mining tasks such as real-time keyword monitoring and unre- stricted vocabulary audio document indexing. However, to date, many keyword spotting approaches have su®ered from poor detection rates, high false alarm rates, or slow execution times, thus reducing their commercial viability. This work investigates the application of keyword spotting to data mining tasks. The thesis makes a number of major contributions to the ¯eld of keyword spotting. The ¯rst major contribution is the development of a novel keyword veri¯cation method named Cohort Word Veri¯cation. This method combines high level lin- guistic information with cohort-based veri¯cation techniques to obtain dramatic improvements in veri¯cation performance, in particular for the problematic short duration target word class. The second major contribution is the development of a novel audio document indexing technique named Dynamic Match Lattice Spotting. This technique aug- ments lattice-based audio indexing principles with dynamic sequence matching techniques to provide robustness to erroneous lattice realisations. The resulting algorithm obtains signi¯cant improvement in detection rate over lattice-based audio document indexing while still maintaining extremely fast search speeds. The third major contribution is the study of multiple veri¯er fusion for the task of keyword veri¯cation. The reported experiments demonstrate that substantial improvements in veri¯cation performance can be obtained through the fusion of multiple keyword veri¯ers. The research focuses on combinations of speech background model based veri¯ers and cohort word veri¯ers. The ¯nal major contribution is a comprehensive study of the e®ects of limited training data for keyword spotting. This study is performed with consideration as to how these e®ects impact the immediate development and deployment of speech technologies for non-English languages.

Validation of the Algase Wandering Scale (Version 2) in a cross cultural sample

This study examined the psychometric properties of an expanded version of the Algase Wandering Scale (Version 2) (AWS-V2) in a cross-cultural sample. A cross-sectional survey design was used. Study subjects were 172 English-speaking persons with dementia (PWD) from long-term care facilities in the USA, Canada, and Australia. Two or more facility staff rated each subject on the AWS-V2. Demographic and cognitive data (MMSE) were also obtained. Staff provided information on their own knowledge of the subject and of dementia. Separate factor analyses on data from two samples of raters each explained greater than 66% of the variance in AWS-V2 scores and validated four (persistent walking, navigational deficit, eloping behavior, and shadowing) of five factors in the original scale. Items added to create the AWS-V2 strengthened the shadowing subscale, failed to improve the routinized walking subscale, and added a factor, attention shifting as compared to the original AWS. Evidence for validity was found in significant correlations and ANOVAs between the AWS-V2 and most subscales with a single item indicator of wandering and with the MMSE. Evidence of reliability was shown by internal consistency of the AWS-V2 (0.87, 0.88) and its subscales (range 0.88 to 0.66), with Kappa for individual items (17 of 27 greater than 0.4), and ANOVAs comparing ratings across rater groups (nurses, nurse aids, and other staff). Analyses support validity and reliability of the AWS-V2 overall and for persistent walking, spatial disorientation, and eloping behavior subscales. The AWS-V2 and its subscales are an appropriate way to measure wandering as conceptualized within the Need-driven Dementia-compromised Behavior Model in studies of English-speaking subjects. Suggestions for further strengthening the scale and for extending its use to clinical applications are described.

Cluster-based network model for time-course gene expression data

We propose a model-based approach to unify clustering and network modeling using time-course gene expression data. Specifically, our approach uses a mixture model to cluster genes. Genes within the same cluster share a similar expression profile. The network is built over cluster-specific expression profiles using state-space models. We discuss the application of our model to simulated data as well as to time-course gene expression data arising from animal models on prostate cancer progression. The latter application shows that with a combined statistical/bioinformatics analyses, we are able to extract gene-to-gene relationships supported by the literature as well as new plausible relationships.

Demographics of international contact lens prescribing

Knowledge of differences in the demographics of contact lens prescribing between nations, and changes over time, can assist (a) the contact lens industry in developing and promoting various product types in different world regions, and (b) practitioners in understanding their prescribing habits in an international context. Data that we have gathered from annual contact lens fitting surveys conducted in Australia, Canada, Japan, the Netherlands, Norway, the UK and the USA between 2000 and 2008 reveal an ageing demographic, with Japan being the most youthful. The majority of fits are to females, with statistically significant differences between nations, ranging from 62 per cent of fits in Norway to 68 per cent in Japan. The small overall decline in the proportion of new fits, and commensurate increase in refits, over the survey periodmay indicate a growing rate of conversion of lens wearers to more advanced lens types, such as silicone hydrogels. � 2009 British Contact Lens Association.

Investigation into travel of transit oriented development : school students’ travel survey of Kelvin Grove Urban Village

Travel surveys were conducted for collecting data related to school students’ travel at Kelvin Grove Urban Village (KGUV). Currently, KGUV has school students studying at grade 10 to 12. As a part of data collection process, travel surveys were undertaken for school students studying. This document contains the questionnaire form used to collect the demographic and travel data related to school students at KGUV. The surveys forms were hand delivered to the school and the responses were collected back via reply paid envelop provided with the questionnaire form.

Encryption safe harbours and data breach notification laws

Data breach notification laws require organisations to notify affected persons or regulatory authorities when an unauthorised acquisition of personal data occurs. Most laws provide a safe harbour to this obligation if acquired data has been encrypted. There are three types of safe harbour: an exemption; a rebuttable presumption and factor-based analysis. We demonstrate, using three condition-based scenarios, that the broad formulation of most encryption safe harbours is based on the flawed assumption that encryption is the silver bullet for personal information protection. We then contend that reliance upon an encryption safe harbour should be dependent upon a rigorous and competent risk-based review that is required on a case-by-case basis. Finally, we recommend the use of both an encryption safe harbour and a notification trigger as our preferred choice for a data breach notification regulatory framework.

Stakeholder perspectives regarding the mandatory notification of Australian data breaches

The advent of data breach notification laws in the United States (US) has unearthed a significant problem involving the mismanagement of personal information by a range of public and private sector organisations. At present, there is currently no statutory obligation under Australian law requiring public or private sector organisations to report a data breach of personal information to law enforcement agencies or affected persons. However, following a comprehensive review of Australian privacy law, the Australian Law Reform Commission (ALRC) has recommended the introduction of a mandatory data breach notification scheme. The issue of data breach notification has ignited fierce debate amongst stakeholders, especially larger private sector entities. The purpose of this article is to document the perspectives of key industry and government representatives to identify their standpoints regarding an appropriate regulatory approach to data breach notification in Australia.

The mandatory notification of data breaches : issues arising for Australian and EU legal developments

Public and private sector organisations are now able to capture and utilise data on a vast scale, thus heightening the importance of adequate measures for protecting unauthorised disclosure of personal information. In this respect, data breach notification has emerged as an issue of increasing importance throughout the world. It has been the subject of law reform in the United States and in other jurisdictions. This article reviews US, Australian and EU legal developments regarding the mandatory notification of data breaches. The authors highlight areas of concern based on the extant US experience that require further consideration in Australia and in the EU.