Decision boundary setting and classifier combination for text classification

This thesis presents a promising boundary setting method for solving challenging issues in text classification to produce an effective text classifier. A classifier must identify boundary between classes optimally. However, after the features are selected, the boundary is still unclear with regard to mixed positive and negative documents. A classifier combination method to boost effectiveness of the classification model is also presented. The experiments carried out in the study demonstrate that the proposed classifier is promising.

Towards making informed decisions on selecting promising work zone safety treatments

Lack of detailed and accurate safety records on incidents in Australian work zones prevents a thorough understanding of the relevant risks and hazards. Consequently it is difficult to select appropriate treatments for improving the safety of roadworkers and motorists alike. This paper outlines development of a conceptual framework for making informed decisions about safety treatments by: 1) identifying safety issues and hazards in work zones; 2) understanding the attitudes and perceptions of both roadworkers and motorists; 3) reviewing the effectiveness of work zone safety treatments according to existing research, and; 4) incorporating local expert opinion on the feasibility and usefulness of the safety treatments. Using data collected through semi-structured interviews with roadwork personnel and online surveys of Queensland drivers, critical safety issues were identified. The effectiveness of treatments for addressing the issues was understood through rigorous literature review and consultations with local road authorities. Promising work zone safety treatments include enforcement, portable rumble strips, perceptual measures to imply reduced lane width, automated or remotely-operated traffic lights, end of queue measures, and more visible and meaningful signage.

Use of online health information to manage children’s health care: A prospective study investigating parental decisions

Background The use of the internet to access information is rapidly increasing; however, the quality of health information provided on various online sites is questionable. We aimed to examine the underlying factors that guide parents' decisions to use online information to manage their child's health care, a behaviour which has not yet been explored systematically. Methods Parents (N=391) completed a questionnaire assessing the standard theory of planned behaviour (TPB) measures of attitude, subjective norm, perceived behavioural control (PBC), and intention as well as the underlying TPB belief-based items (i.e., behavioural, normative, and control beliefs) in addition to a measure of perceived risk and demographic variables. Two months later, consenting parents completed a follow-up telephone questionnaire which assessed the decisions they had made regarding their use of online information to manage their child's health care during the previous 2 months. Results We found support for the TPB constructs of attitude, subjective norm, and PBC as well as the additional construct of perceived risk in predicting parents' intentions to use online information to manage their child's health care, with further support found for intentions, but not PBC, in predicting parents' behaviour. The results of the TPB belief-based analyses also revealed important information about the critical beliefs that guide parents' decisions to engage in this child health management behaviour. Conclusions This theory-based investigation to understand parents' motivations and online information-seeking behaviour is key to developing recommendations and policies to guide more appropriate help-seeking actions among parents.

Pull the plug or take the plunge: Multiple opportunities and the speed of venturing decisions in the Australian mining industry

Effectively capturing opportunities requires rapid decision-making. We investigate the speed of opportunity evaluation decisions by focusing on firms' venture termination and venture advancement decisions. Experience, standard operating procedures, and confidence allow firms to make opportunity evaluation decisions faster; we propose that a firm's attentional orientation, as reflected in its project portfolio, limits the number of domains in which these speed-enhancing mechanisms can be developed. Hence firms' decision speed is likely to vary between different types of decisions. Using unique data on 3,269 mineral exploration ventures in the Australian mining industry, we find that firms with a higher degree of attention toward earlier-stage exploration activities are quicker to abandon potential opportunities in early development but slower to do so later, and that such firms are also slower to advance on potential opportunities at all stages compared to firms that focus their attention differently. Market dynamism moderates these relationships, but only with regard to initial evaluation decisions. Our study extends research on decision speed by showing that firms are not necessarily fast or slow regarding all the decisions they make, and by offering an opportunity evaluation framework that recognizes that decision makers can, in fact often do, pursue multiple potential opportunities simultaneously.

Decisions and actions of distracted drivers at the onset of yellow light

Driving on an approach to a signalized intersection while distracted is relatively risky, as potential vehicular conflicts and resulting angle collisions tend to be relatively more severe compared to other locations. Given the prevalence and importance of this particular scenario, the objective of this study was to examine the decisions and actions of distracted drivers during the onset of yellow lights. Driving simulator data were obtained from a sample of 69 drivers under baseline and handheld cell phone conditions at the University of Iowa – National Advanced Driving Simulator. Explanatory variables included age, gender, cell phone use, distance to stop-line, and speed. Although there is extensive research on drivers’ responses to yellow traffic signals, the examinations have been conducted from a traditional regression-based approach, which do not necessary provide the underlying relations and patterns among the sampled data. In this paper, we exploit the benefits of both classical statistical inference and data mining techniques to identify the a priori relationships among main effects, non-linearities, and interaction effects. Results suggest that the probability of yellow light running increases with the increase in driving speed at the onset of yellow. Both young (18–25 years) and middle-aged (30–45 years) drivers reveal reduced propensity for yellow light running whilst distracted across the entire speed range, exhibiting possible risk compensation during this critical driving situation. The propensity for yellow light running for both distracted male and female older (50–60 years) drivers is significantly higher. Driver experience captured by age interacts with distraction, resulting in their combined effect having slower physiological response and being distracted particularly risky.

Don't argue with the ref! - Legal liability for incorrect decisions of sports' officials

This paper continues the conversation from recent articles examining potential remedies available for incorrect decisions by sports officials. In particular, this article focuses on bringing an action against an official in negligence for pure economic loss. Using precedent cases, it determines that such an action would have a low chance of success, as a duty of care would be difficult to establish. Even if that could be overcome, an aggrieved player or team would still face further hurdles at the stages of breach, causation and defences. The article concludes by proposing some options to further reduce the small risk of liability to officials.

The political decisions and policy leading to the Royal Australian Air Force having no fighters or interceptors for the coming war against Japan

The thesis provides an understanding of the ignored need for a modern air defence system for the Australian air force to meet the growing threat from Japan in the 1930s and early 1940s. The quality of advice provided to, and accepted by, Australian politicians was misleading and eliminated the need for fighters and interceptors despite glaring evidence to the contrary. Based on primary source material, including official documents, Allied and Axis pilot memoirs, popular aviation literature and newspaper and magazine articles and interviews, the thesis highlights the inability of Australian politicians to face the reality of the international situation.

No-reference image quality assessment using modified extreme learning machine classifier

In this paper, we present a machine learning approach to measure the visual quality of JPEG-coded images. The features for predicting the perceived image quality are extracted by considering key human visual sensitivity (HVS) factors such as edge amplitude, edge length, background activity and background luminance. Image quality assessment involves estimating the functional relationship between HVS features and subjective test scores. The quality of the compressed images are obtained without referring to their original images ('No Reference' metric). Here, the problem of quality estimation is transformed to a classification problem and solved using extreme learning machine (ELM) algorithm. In ELM, the input weights and the bias values are randomly chosen and the output weights are analytically calculated. The generalization performance of the ELM algorithm for classification problems with imbalance in the number of samples per quality class depends critically on the input weights and the bias values. Hence, we propose two schemes, namely the k-fold selection scheme (KS-ELM) and the real-coded genetic algorithm (RCGA-ELM) to select the input weights and the bias values such that the generalization performance of the classifier is a maximum. Results indicate that the proposed schemes significantly improve the performance of ELM classifier under imbalance condition for image quality assessment. The experimental results prove that the estimated visual quality of the proposed RCGA-ELM emulates the mean opinion score very well. The experimental results are compared with the existing JPEG no-reference image quality metric and full-reference structural similarity image quality metric.

A randomized trial of fellowships for early career researchers finds a high reliability in funding decisions

Objectives Funding for early career researchers in Australia's largest medical research funding scheme is determined by a competitive peer-review process using a panel of four reviewers. The purpose of this experiment was to appraise the reliability of funding by duplicating applications that were considered by separate grant review panels. Study Design and Methods Sixty duplicate applications were considered by two independent grant review panels that were awarding funding for Australia's National Health and Medical Research Council. Panel members were blinded to which applications were included in the experiment and to whether it was the original or duplicate application. Scores were compared across panels using Bland–Altman plots to determine measures of agreement, including whether agreement would have impacted on actual funding. Results Twenty-three percent of the applicants were funded by both panels and 60 percent were not funded by both, giving an overall agreement of 83 percent [95% confidence interval (CI): 73%, 92%]. The chance-adjusted agreement was 0.75 (95% CI: 0.58, 0.92). Conclusion There was a comparatively high level of agreement when compared with other types of funding schemes. Further experimental research could be used to determine if this higher agreement is due to nature of the application, the composition of the assessment panel, or the characteristics of the applicants.

Speeding up AdaBoost Classifier with Random Projection

The development of techniques for scaling up classifiers so that they can be applied to problems with large datasets of training examples is one of the objectives of data mining. Recently, AdaBoost has become popular among machine learning community thanks to its promising results across a variety of applications. However, training AdaBoost on large datasets is a major problem, especially when the dimensionality of the data is very high. This paper discusses the effect of high dimensionality on the training process of AdaBoost. Two preprocessing options to reduce dimensionality, namely the principal component analysis and random projection are briefly examined. Random projection subject to a probabilistic length preserving transformation is explored further as a computationally light preprocessing step. The experimental results obtained demonstrate the effectiveness of the proposed training process for handling high dimensional large datasets.

Understanding consumers' decisions to adopt technology-enabled transformative services

This study extends understanding of consumers' decisions to adopt transformative services delivered via technology. It incorporates competitive effects into the model of goal-directed behavior which, in keeping with the majority of consumer decision making models, neglects to explicitly account for competition. A goal-level operationalization of competition, incorporating both direct and indirect competition, is proposed. A national web-based survey collected data from 431 respondents about their decisions to adopt mental health services delivered via mobile phone. The findings show that the extent to which consumers perceived using these transformative services to be more instrumental to achieving their goals than competition had the greatest impact on their adoption decisions. This finding builds on the limited empirical evidence for the inclusion of competitive effects to more fully explain consumers' decisions to adopt technology-based and other services. It also provides support for a broader operationalization of competition with respect to consumers' personal goals.

An integrated model of an open-pit coal mine: Improving energy efficiency decisions

This article contributes an original integrated model of an open-pit coal mine for supporting energy-efficient decisions. Mixed integer linear programming is used to formulate a general integrated model of the operational energy consumption of four common open-pit coal mining subsystems: excavation and haulage, stockpiles, processing plants and belt conveyors. Mines are represented as connected instances of the four subsystems, in a flow sheet manner, which are then fitted to data provided by the mine operators. Solving the integrated model ensures the subsystems’ operations are synchronised and whole-of-mine energy efficiency is encouraged. An investigation on a case study of an open-pit coal mine is conducted to validate the proposed methodology. Opportunities are presented for using the model to aid energy-efficient decision-making at various levels of a mine, and future work to improve the approach is described.

Validation-Based Sparse Gaussian Process Classifier Design

Gaussian processes (GPs) are promising Bayesian methods for classification and regression problems. Design of a GP classifier and making predictions using it is, however, computationally demanding, especially when the training set size is large. Sparse GP classifiers are known to overcome this limitation. In this letter, we propose and study a validation-based method for sparse GP classifier design. The proposed method uses a negative log predictive (NLP) loss measure, which is easy to compute for GP models. We use this measure for both basis vector selection and hyperparameter adaptation. The experimental results on several real-world benchmark data sets show better orcomparable generalization performance over existing methods.

Understanding the determinants of Australian hospital nurses' hand hygiene decisions following the implementation of a national hand hygiene initiative

Hand hygiene is the primary measure in hospitals to reduce the spread of infections, with nurses experiencing the greatest frequency of patient contact. The ‘5 critical moments’ of hand hygiene initiative has been implemented in hospitals across Australia, accompanied by awareness-raising, staff training and auditing. The aim of this study was to understand the determinants of nurses’ hand hygiene decisions, using an extension of a common health decision-making model, the theory of planned behaviour (TPB), to inform future health education strategies to increase compliance. Nurses from 50 Australian hospitals (n = 2378) completed standard TPB measures (attitude, subjective norm, perceived behavioural control [PBC], intention) and the extended variables of group norm, risk perceptions (susceptibility, severity) and knowledge (subjective, objective) at Time 1, while a sub-sample (n = 797) reported their hand hygiene behaviour 2 weeks later. Regression analyses identified subjective norm, PBC, group norm, subjective knowledge and risk susceptibility as the significant predictors of nurses’ hand hygiene intentions, with intention and PBC predicting their compliance behaviour. Rather than targeting attitudes which are already very favourable among nurses, health education strategies should focus on normative influences and perceptions of control and risk in efforts to encourage hand hygiene adherence.

The Effect of Framing on Layoff Decisions: Are More-Experienced Managers More Rational? Does Computerized Data Make Any Difference?

[Excerpt] The effects of framing on decisions has been widely studied, producing research that suggests individuals respond to framing in predictable and fairly consistent ways (Bazerman, 1984, 1990; Tversky & Kahneman, 1986; Thaler, 1980). The essential finding from this body of research is that "individuals treat risks concerning perceived gains (for example, saving jobs and plants) differently from risks concerning perceived losses (losing jobs and plants)" (Bazerman, 1990, pp. 49-50). Specifically, individuals tend to avoid risks concerning gains, and seek risks concerning losses.