Towards a reference model for SOA governance (extended version)

Although the lack of elaborate governance mechanisms is often seen as the main reason for failures of SOA projects, SOA governance is still very low in maturity. In this paper, we follow a design science approach to address this drawback by presenting a framework that can guide organisations in implementing a governance approach for SOA more successfully. We have reviewed the highly advanced IT governance frameworks Cobit and ITIL and mapped them to the SOA domain. The resulting blueprint for an SOA governance framework was refined based on a detailed literature review, expert interviews and a practical application in a government organisation. The proposed framework stresses the need for business representatives to get involved in SOA decisions and to define benefits ownership for services.

Managing knowledge transfer as a strategic approach to competitive advantage

Knowledge has been recognised as a source of competitive advantage. Knowledge-based resources allow organisations to adapt products and services to the marketplace and deal with competitive challenges that enable them to compete more effectively. One factor critical to using knowledge-based resources is the ability to transfer knowledge as a dimension of the learning organisation. There are many elements that may influence whether knowledge transfer can be effectively achieved in an organisation such as leadership, problem-solving behaviours, support structures, change management capabilities, absorptive capacity and the nature of the knowledge. An existing framework was applied in a case study to explain how knowledge transfer can be managed effectively and to identify emerging issues or additional factors involved in the process. As a result, a refined framework is proposed that provides a better understanding for the effective management of knowledge transfer processes that can provide a competitive advantage.

Complex modelling of open system design for sustainable architecture

This paper argues a model of complex system design for sustainable architecture within a framework of entropy evolution. The spectrum of sustainable architecture consists of the efficient use of energy and material resource in life-cycle of buildings, the active involvement of the occupants in micro-climate control within buildings, and the natural environmental context. The interactions of the parameters compose a complex system of sustainable architectural design, of which the conventional linear and fragmented design technologies are insufficient to indicate holistic and ongoing environmental performance. The complexity theory of dissipative structure states a microscopic formulation of open system evolution, which provides a system design framework for the evolution of building environmental performance towards an optimization of sustainability in architecture.

Improving automatic speaker verification using SVM techniques

Automatic recognition of people is an active field of research with important forensic and security applications. In these applications, it is not always possible for the subject to be in close proximity to the system. Voice represents a human behavioural trait which can be used to recognise people in such situations. Automatic Speaker Verification (ASV) is the process of verifying a persons identity through the analysis of their speech and enables recognition of a subject at a distance over a telephone channel { wired or wireless. A significant amount of research has focussed on the application of Gaussian mixture model (GMM) techniques to speaker verification systems providing state-of-the-art performance. GMM's are a type of generative classifier trained to model the probability distribution of the features used to represent a speaker. Recently introduced to the field of ASV research is the support vector machine (SVM). An SVM is a discriminative classifier requiring examples from both positive and negative classes to train a speaker model. The SVM is based on margin maximisation whereby a hyperplane attempts to separate classes in a high dimensional space. SVMs applied to the task of speaker verification have shown high potential, particularly when used to complement current GMM-based techniques in hybrid systems. This work aims to improve the performance of ASV systems using novel and innovative SVM-based techniques. Research was divided into three main themes: session variability compensation for SVMs; unsupervised model adaptation; and impostor dataset selection. The first theme investigated the differences between the GMM and SVM domains for the modelling of session variability | an aspect crucial for robust speaker verification. Techniques developed to improve the robustness of GMMbased classification were shown to bring about similar benefits to discriminative SVM classification through their integration in the hybrid GMM mean supervector SVM classifier. Further, the domains for the modelling of session variation were contrasted to find a number of common factors, however, the SVM-domain consistently provided marginally better session variation compensation. Minimal complementary information was found between the techniques due to the similarities in how they achieved their objectives. The second theme saw the proposal of a novel model for the purpose of session variation compensation in ASV systems. Continuous progressive model adaptation attempts to improve speaker models by retraining them after exploiting all encountered test utterances during normal use of the system. The introduction of the weight-based factor analysis model provided significant performance improvements of over 60% in an unsupervised scenario. SVM-based classification was then integrated into the progressive system providing further benefits in performance over the GMM counterpart. Analysis demonstrated that SVMs also hold several beneficial characteristics to the task of unsupervised model adaptation prompting further research in the area. In pursuing the final theme, an innovative background dataset selection technique was developed. This technique selects the most appropriate subset of examples from a large and diverse set of candidate impostor observations for use as the SVM background by exploiting the SVM training process. This selection was performed on a per-observation basis so as to overcome the shortcoming of the traditional heuristic-based approach to dataset selection. Results demonstrate the approach to provide performance improvements over both the use of the complete candidate dataset and the best heuristically-selected dataset whilst being only a fraction of the size. The refined dataset was also shown to generalise well to unseen corpora and be highly applicable to the selection of impostor cohorts required in alternate techniques for speaker verification.

Model round-trip engineering

In the quest for shorter time-to-market, higher quality and reduced cost, model-driven software development has emerged as a promising approach to software engineering. The central idea is to promote models to first-class citizens in the development process. Starting from a set of very abstract models in the early stage of the development, they are refined into more concrete models and finally, as a last step, into code. As early phases of development focus on different concepts compared to later stages, various modelling languages are employed to most accurately capture the concepts and relations under discussion. In light of this refinement process, translating between modelling languages becomes a time-consuming and error-prone necessity. This is remedied by model transformations providing support for reusing and automating recurring translation efforts. These transformations typically can only be used to translate a source model into a target model, but not vice versa. This poses a problem if the target model is subject to change. In this case the models get out of sync and therefore do not constitute a coherent description of the software system anymore, leading to erroneous results in later stages. This is a serious threat to the promised benefits of quality, cost-saving, and time-to-market. Therefore, providing a means to restore synchronisation after changes to models is crucial if the model-driven vision is to be realised. This process of reflecting changes made to a target model back to the source model is commonly known as Round-Trip Engineering (RTE). While there are a number of approaches to this problem, they impose restrictions on the nature of the model transformation. Typically, in order for a transformation to be reversed, for every change to the target model there must be exactly one change to the source model. While this makes synchronisation relatively “easy”, it is ill-suited for many practically relevant transformations as they do not have this one-to-one character. To overcome these issues and to provide a more general approach to RTE, this thesis puts forward an approach in two stages. First, a formal understanding of model synchronisation on the basis of non-injective transformations (where a number of different source models can correspond to the same target model) is established. Second, detailed techniques are devised that allow the implementation of this understanding of synchronisation. A formal underpinning for these techniques is drawn from abductive logic reasoning, which allows the inference of explanations from an observation in the context of a background theory. As non-injective transformations are the subject of this research, there might be a number of changes to the source model that all equally reflect a certain target model change. To help guide the procedure in finding “good” source changes, model metrics and heuristics are investigated. Combining abductive reasoning with best-first search and a “suitable” heuristic enables efficient computation of a number of “good” source changes. With this procedure Round-Trip Engineering of non-injective transformations can be supported.

Data-driven background dataset selection for SVM-based speaker verification

The recently proposed data-driven background dataset refinement technique provides a means of selecting an informative background for support vector machine (SVM)-based speaker verification systems. This paper investigates the characteristics of the impostor examples in such highly-informative background datasets. Data-driven dataset refinement individually evaluates the suitability of candidate impostor examples for the SVM background prior to selecting the highest-ranking examples as a refined background dataset. Further, the characteristics of the refined dataset were analysed to investigate the desired traits of an informative SVM background. The most informative examples of the refined dataset were found to consist of large amounts of active speech and distinctive language characteristics. The data-driven refinement technique was shown to filter the set of candidate impostor examples to produce a more disperse representation of the impostor population in the SVM kernel space, thereby reducing the number of redundant and less-informative examples in the background dataset. Furthermore, data-driven refinement was shown to provide performance gains when applied to the difficult task of refining a small candidate dataset that was mis-matched to the evaluation conditions.

Exploiting multiple feature sets in data-driven impostor dataset selection for speaker verification

This study assesses the recently proposed data-driven background dataset refinement technique for speaker verification using alternate SVM feature sets to the GMM supervector features for which it was originally designed. The performance improvements brought about in each trialled SVM configuration demonstrate the versatility of background dataset refinement. This work also extends on the originally proposed technique to exploit support vector coefficients as an impostor suitability metric in the data-driven selection process. Using support vector coefficients improved the performance of the refined datasets in the evaluation of unseen data. Further, attempts are made to exploit the differences in impostor example suitability measures from varying features spaces to provide added robustness.

Tax morale and conditional cooperation

Why so many people pay their taxes, even though fines and audit probability are low, is a central question in the tax compliance literature. Positing a homo oeconomicus having a refined motivation structure sheds light on this puzzle. This paper provides empirical evidence for the relevance of conditional cooperation, using survey data from 30 West and East European countries. We find a high correlation between perceived tax evasion and tax morale. The results remain robust after exploiting endogeneity and conducting several robustness tests. We also observe a strong positive correlation between institutional quality and tax mmorale. Keywords: Tax morale; Tax compliance; Tax evasion; Pro-social behavior; Institutions

Development of a Systematic Approach to Assessing Quality within Australian Residential Aged Care Facilities: The Clinical Care Indicators Tool

Recent years have seen the introduction of formalised accreditation processes in both community and residential aged care, but these only partially address quality assessment within this sector. Residential aged care in Australia does not yet have a standardised system of resident assessment related to clinical, rather than administrative, outcomes. This paper describes the development of a quality assessment tool aimed at addressing this gap. Utilising previous research and the results of nominal groups with experts in the field, the 21-item Clinical Care Indicators (CCI) Tool for residential aged care was developed and trialled nationally. The CCI Tool was found to be simple to use and an effective means of collecting data on the state of resident health and care, with potential benefits for resident care planning and continuous quality improvement within facilities and organisations. The CCI Tool was further refined through a small intervention study to assess its utility as a quality improvement instrument and to investigate its relationship with resident quality of life. The current version covers 23 clinical indicators, takes about 30 minutes to complete and is viewed favourably by nursing staff who use it. Current work focuses on psychometric analysis and benchmarking, which should enable the CCI Tool to make a positive contribution to the measurement of quality in aged care in Australia.

Predicting loss and fragmentation of habitat of the vulnerable subtropical rainforest tree Macadamia integrifolia with models developed from compiled ecological data

Habitat models are widely used in ecology, however there are relatively few studies of rare species, primarily because of a paucity of survey records and lack of robust means of assessing accuracy of modelled spatial predictions. We investigated the potential of compiled ecological data in developing habitat models for Macadamia integrifolia, a vulnerable mid-stratum tree endemic to lowland subtropical rainforests of southeast Queensland, Australia. We compared performance of two binomial models—Classification and Regression Trees (CART) and Generalised Additive Models (GAM)—with Maximum Entropy (MAXENT) models developed from (i) presence records and available absence data and (ii) developed using presence records and background data. The GAM model was the best performer across the range of evaluation measures employed, however all models were assessed as potentially useful for informing in situ conservation of M. integrifolia, A significant loss in the amount of M. integrifolia habitat has occurred (p < 0.05), with only 37% of former habitat (pre-clearing) remaining in 2003. Remnant patches are significantly smaller, have larger edge-to-area ratios and are more isolated from each other compared to pre-clearing configurations (p < 0.05). Whilst the network of suitable habitat patches is still largely intact, there are numerous smaller patches that are more isolated in the contemporary landscape compared with their connectedness before clearing. These results suggest that in situ conservation of M. integrifolia may be best achieved through a landscape approach that considers the relative contribution of small remnant habitat fragments to the species as a whole, as facilitating connectivity among the entire network of habitat patches.

Predicting active material utilisation in LiFePO4 electrodes using a multi-scale mathematical model

A mathematical model is developed to simulate the discharge of a LiFePO4 cathode. This model contains 3 size scales, which match with experimental observations present in the literature on the multi-scale nature of LiFePO4 material. A shrinking-core is used on the smallest scale to represent the phase-transition of LiFePO4 during discharge. The model is then validated against existing experimental data and this validated model is then used to investigate parameters that influence active material utilisation. Specifically the size and composition of agglomerates of LiFePO4 crystals is discussed, and we investigate and quantify the relative effects that the ionic and electronic conductivities within the oxide have on oxide utilisation. We find that agglomerates of crystals can be tolerated under low discharge rates. The role of the electrolyte in limiting (cathodic) discharge is also discussed, and we show that electrolyte transport does limit performance at high discharge rates, confirming the conclusions of recent literature.