Legislative and security requirements of audit material for evidentiary purpose

This research used the Queensland Police Service, Australia, as a major case study. Information on principles, techniques and processes used, and the reason for the recording, storing and release of audit information for evidentiary purposes is reported. It is shown that Law Enforcement Agencies have a two-fold interest in, and legal obligation pertaining to, audit trails. The first interest relates to the situation where audit trails are actually used by criminals in the commission of crime and the second to where audit trails are generated by the information systems used by the police themselves in support of the recording and investigation of crime. Eleven court cases involving Queensland Police Service audit trails used in evidence in Queensland courts were selected for further analysis. It is shown that, of the cases studied, none of the evidence presented was rejected or seriously challenged from a technical perspective. These results were further analysed and related to normal requirements for trusted maintenance of audit trail information in sensitive environments with discussion on the ability and/or willingness of courts to fully challenge, assess or value audit evidence presented. Managerial and technical frameworks for firstly what is considered as an environment where a computer system may be considered to be operating “properly” and, secondly, what aspects of education, training, qualifications, expertise and the like may be considered as appropriate for persons responsible within that environment, are both proposed. Analysis was undertaken to determine if audit and control of information in a high security environment, such as law enforcement, could be judged as having improved, or not, in the transition from manual to electronic processes. Information collection, control of processing and audit in manual processes used by the Queensland Police Service, Australia, in the period 1940 to 1980 was assessed against current electronic systems essentially introduced to policing in the decades of the 1980s and 1990s. Results show that electronic systems do provide for faster communications with centrally controlled and updated information readily available for use by large numbers of users who are connected across significant geographical locations. However, it is clearly evident that the price paid for this is a lack of ability and/or reluctance to provide improved audit and control processes. To compare the information systems audit and control arrangements of the Queensland Police Service with other government departments or agencies, an Australia wide survey was conducted. Results of the survey were contrasted with the particular results of a survey, conducted by the Australian Commonwealth Privacy Commission four years previous, to this survey which showed that security in relation to the recording of activity against access to information held on Australian government computer systems has been poor and a cause for concern. However, within this four year period there is evidence to suggest that government organisations are increasingly more inclined to generate audit trails. An attack on the overall security of audit trails in computer operating systems was initiated to further investigate findings reported in relation to the government systems survey. The survey showed that information systems audit trails in Microsoft Corporation's “Windows” operating system environments are relied on quite heavily. An audit of the security for audit trails generated, stored and managed in the Microsoft “Windows 2000” operating system environment was undertaken and compared and contrasted with similar such audit trail schemes in the “UNIX” and “Linux” operating systems. Strength of passwords and exploitation of any security problems in access control were targeted using software tools that are freely available in the public domain. Results showed that such security for the “Windows 2000” system is seriously flawed and the integrity of audit trails stored within these environments cannot be relied upon. An attempt to produce a framework and set of guidelines for use by expert witnesses in the information technology (IT) profession is proposed. This is achieved by examining the current rules and guidelines related to the provision of expert evidence in a court environment, by analysing the rationale for the separation of distinct disciplines and corresponding bodies of knowledge used by the Medical Profession and Forensic Science and then by analysing the bodies of knowledge within the discipline of IT itself. It is demonstrated that the accepted processes and procedures relevant to expert witnessing in a court environment are transferable to the IT sector. However, unlike some discipline areas, this analysis has clearly identified two distinct aspects of the matter which appear particularly relevant to IT. These two areas are; expertise gained through the application of IT to information needs in a particular public or private enterprise; and expertise gained through accepted and verifiable education, training and experience in fundamental IT products and system.

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.