New variational Bayesian approaches for statistical data mining : with applications to profiling and differentiating habitual consumption behaviour of customers in the wireless telecommunication industry

This thesis investigates profiling and differentiating customers through the use of statistical data mining techniques. The business application of our work centres on examining individuals’ seldomly studied yet critical consumption behaviour over an extensive time period within the context of the wireless telecommunication industry; consumption behaviour (as oppose to purchasing behaviour) is behaviour that has been performed so frequently that it become habitual and involves minimal intentions or decision making. Key variables investigated are the activity initialised timestamp and cell tower location as well as the activity type and usage quantity (e.g., voice call with duration in seconds); and the research focuses are on customers’ spatial and temporal usage behaviour. The main methodological emphasis is on the development of clustering models based on Gaussian mixture models (GMMs) which are fitted with the use of the recently developed variational Bayesian (VB) method. VB is an efficient deterministic alternative to the popular but computationally demandingMarkov chainMonte Carlo (MCMC) methods. The standard VBGMMalgorithm is extended by allowing component splitting such that it is robust to initial parameter choices and can automatically and efficiently determine the number of components. The new algorithm we propose allows more effective modelling of individuals’ highly heterogeneous and spiky spatial usage behaviour, or more generally human mobility patterns; the term spiky describes data patterns with large areas of low probability mixed with small areas of high probability. Customers are then characterised and segmented based on the fitted GMM which corresponds to how each of them uses the products/services spatially in their daily lives; this is essentially their likely lifestyle and occupational traits. Other significant research contributions include fitting GMMs using VB to circular data i.e., the temporal usage behaviour, and developing clustering algorithms suitable for high dimensional data based on the use of VB-GMM.

Which older women could benefit from interventions to decrease sitting time and increase physical activity?

In addition to the well-known health risks associated with lack of physical activity (PA), evidence is emerging about the health risks of sedentary behaviour (sitting). Research about patterns and correlates of sitting and PA in older women is scarce. METHODS: Self-report data from 6,116 women aged 76-81 years were collected as part of the Australian Longitudinal Study on Woman’s Health. Linear regression models were computed to examine whether demographic, social and health factors were associated with sitting and PA. RESULTS: Women who did no PA sat more than women who did any PA (p<0.001). Seven correlates were associated with sitting and PA (p<0.05). Five of these were associated with more sitting and less PA: three health-related (BMI, chronic conditions, anxiety/depression) and two social correlates (caring duties, volunteering). One demographic (being from another English-speaking country) and one social correlate (more social interaction) were associated with more sitting and more PA. Four correlates, two demographic (living in a city; post-high school education), one social (being single), and one health-related correlate (dizziness/loss of balance) were associated with more sitting only. Two other health-related correlates (stiff/painful joints; feet problems) were associated with less PA only. CONCLUSION: Sedentary behaviour and PA are distinct behaviours in older Australian women. Information about the correlates of both behaviours can be used to identify population groups who might benefit from interventions to reduce sedentary behaviour and/or increase PA.

The association between objectively measured neighborhood features and walking in middle-aged adults

Purpose: To explore the role of the neighborhood environment in supporting walking Design: Cross sectional study of 10,286 residents of 200 neighborhoods. Participants were selected using a stratified two-stage cluster design. Data were collected by mail survey (68.5% response rate). Setting: The Brisbane City Local Government Area, Australia, 2007. Subjects: Brisbane residents aged 40 to 65 years. Measures Environmental: street connectivity, residential density, hilliness, tree coverage, bikeways, and street lights within a one kilometer circular buffer from each resident’s home; and network distance to nearest river or coast, public transport, shop, and park. Walking: minutes in the previous week categorized as < 30 minutes, ≥ 30 < 90 minutes, ≥ 90 < 150 minutes, ≥ 150 < 300 minutes, and ≥ 300 minutes. Analysis: The association between each neighborhood characteristic and walking was examined using multilevel multinomial logistic regression and the model parameters were estimated using Markov chain Monte Carlo simulation. Results: After adjustment for individual factors, the likelihood of walking for more than 300 minutes (relative to <30 minutes) was highest in areas with the most connectivity (OR=1.93, 99% CI 1.32-2.80), the greatest residential density (OR=1.47, 99% CI 1.02-2.12), the least tree coverage (OR=1.69, 99% CI 1.13-2.51), the most bikeways (OR=1.60, 99% CI 1.16-2.21), and the most street lights (OR=1.50, 99% CI 1.07-2.11). The likelihood of walking for more than 300 minutes was also higher among those who lived closest to a river or the coast (OR=2.06, 99% CI 1.41-3.02). Conclusion: The likelihood of meeting (and exceeding) physical activity recommendations on the basis of walking was higher in neighborhoods with greater street connectivity and residential density, more street lights and bikeways, closer proximity to waterways, and less tree coverage. Interventions targeting these neighborhood characteristics may lead to improved environmental quality as well as lower rates of overweight and obesity and associated chromic disease.

Principles of Criminal Law in Queensland and Western Australia

Aspires to engage students in the criminal law in Queensland and Western Australia by explaining and applying its fundamental principles. In both these jurisdictions the criminal law stems from the Griffith Code, which was enacted in 1899 in Queensland and in 1913 in Western Australia.

Sitting-time, physical activity and depressive symptoms in mid-aged women

Background: Associations between sitting-time and physical activity (PA) with depression are unclear. Purpose: To examine concurrent and prospective associations between both sitting-time and PA with prevalent depressive symptoms in mid-aged Australian women. Methods: Data were from 8,950 women, aged 50-55 years in 2001, who completed mail surveys in 2001, 2004, 2007 and 2010. Depressive symptoms were assessed using the Center for Epidemiological Studies Depression questionnaire. Associations between sitting-time (≤4, >4-7, >7 hrs/day) and PA (none, some, meeting guidelines) with depressive symptoms (symptoms/no symptoms) were examined in 2011 in concurrent and lagged mixed effect logistic modeling. Both main effects and interaction models were developed. Results: In main effects modeling, women who sat >7 hrs/day (OR 1.47, 95%CI 1.29-1.67) and women who did no PA (OR 1.99, 95%CI 1.75-2.27) were more likely to have depressive symptoms than women who sat ≤4 hrs/day and who met PA guidelines, respectively. In interaction modeling, the likelihood of depressive symptoms in women who sat >7 hrs/day and did no PA was triple that of women who sat ≤4 hrs/day and met PA guidelines (OR 2.96, 95%CI 2.37-3.69). In prospective main effects and interaction modeling, sitting-time was not associated with depressive symptoms, but women who did no PA were more likely than those who met PA guidelines to have future depressive symptoms (OR 1.26, 95%CI 1.08-1.47). Conclusions: Increasing PA to a level commensurate with PA guidelines can alleviate current depression symptoms and prevent future symptoms in mid-aged women. Reducing sitting-time may ameliorate current symptoms.

A retrospective cohort investigation of active range of motion within one week of open reduction and internal fixation of distal radius fractures

Distal radius fractures stabilized by open reduction internal fixation (ORIF) have become increasingly common. There is currently no consensus on the optimal time to commence range of motion (ROM) exercises post-ORIF. A retrospective cohort review was conducted over a five-year period to compare wrist and forearm range of motion outcomes and number of therapy sessions between patients who commenced active ROM exercises within the first seven days and from day eight onward following ORIF of distal radius fractures. One hundred and twenty-one patient cases were identified. Clinical data, active ROM at initial and discharge therapy assessments, fracture type, surgical approaches, and number of therapy sessions attended were recorded. One hundred and seven (88.4%) cases had complete datasets. The early active ROM group (n = 37) commenced ROM a mean (SD) of 4.27 (1.8) days post-ORIF. The comparator group (n = 70) commenced ROM exercises 24.3 (13.6) days post-ORIF. No significant differences were identified between groups in ROM at initial or discharge assessments, or therapy sessions attended. The results from this study indicate that patients who commenced active ROM exercises an average of 24 days after surgery achieved comparable ROM outcomes with similar number of therapy sessions to those who commenced ROM exercises within the first week.

Rethinking disaster risk management and climate change adaptation

Australian governments face the twin challenges of dealing with extreme weather-related disasters (such as floods and bushfires) and adapting to the impacts of climate change. These challenges are connected, so any response would benefit from a more integrated approach across and between the different levels of government.This report summarises the findings of an NCCARF-funded project that addresses this problem. The project undertook a three-way comparative case study of the 2009 Victorian bushfires, the 2011 Perth Hills bushfires, and the 2011 Brisbane floods. It collected data from the official inquiry reports into each of these events, and conducted new interviews and workshops with key stakeholders. The findings of this project included recommendations that range from the conceptual to the practical. First, it was argued that a reconceptualization of terms such as ‘community’ and ‘resilience’ was necessary to allow for more tailored responses to varying circumstances. Second, it was suggested that the high level of uncertainty inherent in disaster risk management and climate change adaptation requires a more iterative approach to policymaking and planning. Third, some specific institutional reforms were proposed that included: 1) a new funding mechanism that would encourage collaboration between and across different levels of government, as well as promoting partnerships with business and the community; 2) improving community engagement through new resilience grants run by local councils; 3) embedding climate change researchers within disaster risk management agencies to promote institutional learning, and; 4) creating an inter-agency network that encourages collaboration between organisations.

The challenge of integrating climate change adaptation and disaster risk management : lessons from bushfire and flood inquiries in an Australian context

Emergency management and climate change adaptation will increasingly challenge all levels of government because of three main factors. First, Australia is extremely vulnerable to the impacts of climate change, particularly through the increasing frequency, duration and/or intensity of disasters such as floods and bushfires. Second, the system of government that divides powers by function and level can often act as a barrier to a well-integrated response. Third, policymaking processes struggle to cope with such complex inter-jurisdictional issues. This paper discusses these factors and explores the nature of the challenge for Australian governments. Investigations into the 2009 Victorian bushfires, the 2011 Perth Hills bushfires, and the 2011 Brisbane floods offer an indication of the challenges ahead and it is argued that there is a need to: improve community engagement and communication; refocus attention on resilience; improve interagency communication and collaboration; and, develop institutional arrangements that support continual improvement and policy learning. These findings offer an opportunity for improving responses as well as a starting point for integrating disaster risk management and climate change adaptation policies. The paper is based on the preliminary findings of an NCCARF funded research project: The Right Tool for the Job: Achieving climate change adaptation outcomes through improved disaster management policies, planning and risk management strategies involving Griffith University and RMIT. It should be noted from the outset that the purpose of this research project is not to criticise the actions of emergency service workers and volunteers who do an incredible job under extreme circumstances, often risking their own lives in the process. The aim is simply to offer emergency management agencies the opportunity to step back and rethink their overall approach to the challenge they face in the light of the impacts of climate change.

Public policy, disaster risk management and climate change adaptation

Public policymakers are caught in a dilemma : there is a growing list of urgent issues to address, at the same time that public expenditure is being cut. Adding to this dilemma is a system of government designed in the 19th century and competing theories of policymaking dating back to the 1950s. The interlinked problems of disaster risk management and climate change adaptation are cases in point. As the climate changes, there will be more frequent, intense and/or prolonged disasters such as floods and bushfires. Clearly a well integrated whole of government response is needed, but how might this be achieved? Further, how could academic research contribute to resolving this dilemma in a way that would produce something of theoretical interest as well as practical outcomes for policymakers? These are the questions addressed by our research via a comparative analysis of the 2009 Victorian bushfires, the 2011 Perth Hills bushfires, and the 2011 Brisbane floods. Our findings suggest that there is a need to: improve community engagement and communication; refocus attention on resilience; improve interagency communication and collaboration; and, develop institutional arrangements that support continual improvement and policy learning. These findings have implications for all areas of public policy theory and practice.