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.

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.

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.

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.

Grassland management activity data : current sources and future needs

Estimates of potential and actual C sequestration require areal information about various types of management activities. Forest surveys, land use data, and agricultural statistics contribute information enabling calculation of the impacts of current and historical land management on C sequestration in biomass (in forests) or in soil (in agricultural systems). Unfortunately little information exists on the distribution of various management activities that can impact soil C content in grassland systems. Limited information of this type restricts our ability to carry out bottom-up estimates of the current C balance of grasslands or to assess the potential for grasslands to act as C sinks with changes in management. Here we review currently available information about grassland management, how that information could be related to information about the impacts of management on soil C stocks, information that may be available in the future, and needs that remain to be filled before in-depth assessments may be carried out. We also evaluate constraints induced by variability in information sources within and between countries. It is readily apparent that activity data for grassland management is collected less frequently and on a coarser scale than data for forest or agricultural inventories and that grassland activity data cannot be directly translated into IPCC-type factors as is done for IPCC inventories of agricultural soils. However, those management data that are available can serve to delineate broad-scale differences in management activities within regions in which soil C is likely to change in response to changes in management. This, coupled with the distinct possibility of more intensive surveys planned in the future, may enable more accurate assessments of grassland C dynamics with higher resolution both spatially and in the number management activities.