Learning the learning rate for prediction with expert advice

Most standard algorithms for prediction with expert advice depend on a parameter called the learning rate. This learning rate needs to be large enough to fit the data well, but small enough to prevent overfitting. For the exponential weights algorithm, a sequence of prior work has established theoretical guarantees for higher and higher data-dependent tunings of the learning rate, which allow for increasingly aggressive learning. But in practice such theoretical tunings often still perform worse (as measured by their regret) than ad hoc tuning with an even higher learning rate. To close the gap between theory and practice we introduce an approach to learn the learning rate. Up to a factor that is at most (poly)logarithmic in the number of experts and the inverse of the learning rate, our method performs as well as if we would know the empirically best learning rate from a large range that includes both conservative small values and values that are much higher than those for which formal guarantees were previously available. Our method employs a grid of learning rates, yet runs in linear time regardless of the size of the grid.

Using viewer’s facial expression and heart rate for sports video highlights detection

Viewer interests, evoked by video content, can potentially identify the highlights of the video. This paper explores the use of facial expressions (FE) and heart rate (HR) of viewers captured using camera and non-strapped sensor for identifying interesting video segments. The data from ten subjects with three videos showed that these signals are viewer dependent and not synchronized with the video contents. To address this issue, new algorithms are proposed to effectively combine FE and HR signals for identifying the time when viewer interest is potentially high. The results show that, compared with subjective annotation and match report highlights, ‘non-neutral’ FE and ‘relatively higher and faster’ HR is able to capture 60%-80% of goal, foul, and shot-on-goal soccer video events. FE is found to be more indicative than HR of viewer’s interests, but the fusion of these two modalities outperforms each of them.

Experimental and numerical investigation of strain-rate dependent mechanical properties of single living cells

The objective of this project is to investigate the strain-rate dependent mechanical behaviour of single living cells using both experimental and numerical techniques. The results revealed that living cells behave as porohyperlastic materials and that both solid and fluid phases within the cells play important roles in their mechanical responses. The research reported in this thesis provides a better understanding of the mechanisms underlying the cellular responses to external mechanical loadings and of the process of mechanical signal transduction in living cells. It would help us to enhance knowledge of and insight into the role of mechanical forces in supporting tissue regeneration or degeneration.

Is group more cost effective than individual cognitive behaviour therapy? The evidence is not solid yet

This paper critically evaluates the empirical evidence of 36 studies regarding the comparative cost-effectiveness of group and individual cognitive behaviour therapy (CBT) as a whole, and also for specific mental disorders (e.g. depression, anxiety, substance abuse) or populations (e.g. children). Methods of calculating costs, as well as methods of comparing treatment outcomes were appraised and criticized. Overall, the evidence that group CBT is more cost-effective than individual CBT is mixed, with group CBT appearing to be more cost effective in treating depression and children, but less cost effective in treating drugs and alcohol dependence, anxiety and social phobias. In addition, methodological weaknesses in the studies assessed are noted. There is a need to improve cost calculation methodology, as well as more solid and a greater number of empirical cost-effectiveness studies before a firm conclusion can be reached that group CBT is more cost effective then individual CBT.

The culture of young women's drinking: Implications for developing effective policing strategies. Report to National Drug Strategy Law Enforcement Committee

Alcohol-related mortality and morbidity represents a substantial financial burden on communities across the world. Adolescence and young adulthood is a peak period for heavy episodic alcohol consumption, with over a third of all people aged 14-19 years having been at risk of acute alcoholrelated harm at least once in the previous 12 months (Australian Institute of Health and Welfare [AIHW], 2011). Excessive alcohol consumption has long been seen as a male problem; however, a gradual shift towards a social acceptance of female drunkenness has narrowed the gap in drinking quantity and style between men and women (Grucza, Bucholz, Rice, & Bierut, 2008). The presented data point to the vulnerability of women to the consequences of acute alcohol intoxication and indicate that alcohol-related offending by women is on the rise. Taken together, these findings reveal that alcohol-related harms and aggression for young women are becoming more prevalent and problematic. This report addressed these issues from a policing perspective...

Key beliefs of hospital nurses’ hand-hygiene behaviour: Protecting your peers and needing effective reminders

Issues addressed: Hand hygiene in hospitals is vital to limit the spread of infections. This study aimed to identify key beliefs underlying hospital nurses’ hand-hygiene decisions to consolidate strategies that encourage compliance. Methods: Informed by a theory of planned behaviour belief framework, nurses from 50 Australian hospitals (n = 797) responded to how likely behavioural beliefs (advantages and disadvantages), normative beliefs (important referents) and control beliefs (barriers) impacted on their hand-hygiene decisions following the introduction of a national ‘5 moments for hand hygiene’ initiative. Two weeks after completing the survey, they reported their hand-hygiene adherence. Stepwise regression analyses identified key beliefs that determined nurses’ hand-hygiene behaviour. Results: Reducing the chance of infection for co-workers influenced nurses’ hygiene behaviour, with lack of time and forgetfulness identified as barriers. Conclusions: Future efforts to improve hand hygiene should highlight the potential impact on colleagues and consider strategies to combat time constraints, as well as implementing workplace reminders to prompt greater hand-hygiene compliance. So what? Rather than emphasising the health of self and patients in efforts to encourage hand-hygiene practices, a focus on peer protection should be adopted and more effective workplace reminders should be implemented to combat forgetting.

Achievement unlocked: Investigating the design of effective gamification experiences for mobile applications and devices

This thesis investigates the design of motivating and engaging software experiences. In particular it examines the use of video game elements in non-game contexts, known as gamification, and how to effectively design gamification experiences for smartphone applications. The original contribution of this thesis is a novel framework for designing gamification, derived from an iterative process of evaluating gamified prototypes. The outcomes of this research can help us to better understand the impact of gamification in today's society and how it can be used to design more effective software.

Investigating a model of mentoring for effective teaching

This paper investigates a five-factor model of mentoring for effective teaching. A survey was administered to 218 student teachers after student teaching to provide insights into their mentoring experience. Results indicated the five factors, namely, personal attributes, system requirements, pedagogical knowledge, modeling, and feedback, had Cronbach alpha scores of .93, .81, .95, .91, and .91, respectively with mean scale scores ranging from 4.20 to 4.60 (p< .001). Items associated with each factor were analyzed; the lowest percentage response was reviewing lesson plans (71%) and the highest percentage was modeling effective teaching practices (96%). Triangulated data from the survey results suggested that the practices implemented by the mentor teachers were perceived to have supported the student teachers’ development during student teaching. Implications of this study suggest that actively engaging mentor teachers who apply the principles outlined by the five factor areas will serve to ensure highly effective support for the development of student teachers.

Centroid training to achieve effective text classification

Traditional text classification technology based on machine learning and data mining techniques has made a big progress. However, it is still a big problem on how to draw an exact decision boundary between relevant and irrelevant objects in binary classification due to much uncertainty produced in the process of the traditional algorithms. The proposed model CTTC (Centroid Training for Text Classification) aims to build an uncertainty boundary to absorb as many indeterminate objects as possible so as to elevate the certainty of the relevant and irrelevant groups through the centroid clustering and training process. The clustering starts from the two training subsets labelled as relevant or irrelevant respectively to create two principal centroid vectors by which all the training samples are further separated into three groups: POS, NEG and BND, with all the indeterminate objects absorbed into the uncertain decision boundary BND. Two pairs of centroid vectors are proposed to be trained and optimized through the subsequent iterative multi-learning process, all of which are proposed to collaboratively help predict the polarities of the incoming objects thereafter. For the assessment of the proposed model, F1 and Accuracy have been chosen as the key evaluation measures. We stress the F1 measure because it can display the overall performance improvement of the final classifier better than Accuracy. A large number of experiments have been completed using the proposed model on the Reuters Corpus Volume 1 (RCV1) which is important standard dataset in the field. The experiment results show that the proposed model has significantly improved the binary text classification performance in both F1 and Accuracy compared with three other influential baseline models.

Development of a cost-effective and flexible vibration DAQ system for long-term continuous structural health monitoring

In the structural health monitoring (SHM) field, long-term continuous vibration-based monitoring is becoming increasingly popular as this could keep track of the health status of structures during their service lives. However, implementing such a system is not always feasible due to on-going conflicts between budget constraints and the need of sophisticated systems to monitor real-world structures under their demanding in-service conditions. To address this problem, this paper presents a comprehensive development of a cost-effective and flexible vibration DAQ system for long-term continuous SHM of a newly constructed institutional complex with a special focus on the main building. First, selections of sensor type and sensor positions are scrutinized to overcome adversities such as low-frequency and low-level vibration measurements. In order to economically tackle the sparse measurement problem, a cost-optimized Ethernet-based peripheral DAQ model is first adopted to form the system skeleton. A combination of a high-resolution timing coordination method based on the TCP/IP command communication medium and a periodic system resynchronization strategy is then proposed to synchronize data from multiple distributed DAQ units. The results of both experimental evaluations and experimental–numerical verifications show that the proposed DAQ system in general and the data synchronization solution in particular work well and they can provide a promising cost-effective and flexible alternative for use in real-world SHM projects. Finally, the paper demonstrates simple but effective ways to make use of the developed monitoring system for long-term continuous structural health evaluation as well as to use the instrumented building herein as a multi-purpose benchmark structure for studying not only practical SHM problems but also synchronization related issues.

Oral liquid nutritional supplements for people with dementia in residential aged care facilities

AIM: This systematic review investigated the prescription, administration and effectiveness of oral liquid nutritional supplements (OLNS) for people with dementia in residential aged care facilities (RACF). METHODS: A comprehensive search of relevant databases, hand searching and cross-referencing found 15 relevant articles from a total of 2910 possible results. Articles which met the inclusion criteria were critically appraised by two independent reviewers using the relevant Joanna Briggs Institute (JBI) appraisal checklist. Data were extracted using the relevant JBI extraction instruments. No data synthesis was possible due to clinical and methodological heterogeneity. RESULTS: Included studies examined a range of strategies, issues and results related to OLNS for persons with dementia in RACFs; however there appear to be significant gaps in the current body of research, particularly in relation to examinations of effectiveness. CONCLUSIONS: This review was unable to produce a definitive finding regarding effectiveness. OLNS may improve the nutritional state of residents with dementia and help prevent weight loss, and there is some suggestion that it may slow the rate of cognitive decline. However, in order for OLNS to be effective, nursing and care staff need to ensure that sufficient attention is paid to the issues of prescription and administration.

Effective removal of zinc (II) from aqueous solutions by tricalcium aluminate (C3A)

Recently, studies have identified high zinc levels in various environmental resources, and excessive intake of zinc has long been considered to be harmful to human health. The aim of this research was to investigate the effectiveness of tricalcium aluminate (C3A) as a removal agent of zinc from aqueous solution. Inductively coupled plasma-atomic emission spectrometer (ICP-AES), X-ray diffraction (XRD) and scanning electron microscopy (SEM) have been used to characterize such removal behavior. The effects of various factors such as pH influence, temperature and contact time were investigated. The adsorption capacity of C3A for Zn2+ was computed to be up to 13.73 mmol g−1, and the highest zinc removal capacity was obtained when the initial pH of Zn(NO3)2 solution was between 6.0 and 7.0, with temperature around 308 K. The XRD analysis showed that the resultant products were ZnAl-LDHs. Combined with the analysis of solution component, it was proved the existence of both precipitation and cation exchange in the removal process. From the experimental results, it was clear that C3A could be potentially used as a cost-effective material for the removal of zinc in aqueous environment.

Data driven and practice-based evidence : design and development of efficient and effective clinical decision support system

Decision-making is such an integral aspect in health care routine that the ability to make the right decisions at crucial moments can lead to patient health improvements. Evidence-based practice, the paradigm used to make those informed decisions, relies on the use of current best evidence from systematic research such as randomized controlled trials. Limitations of the outcomes from randomized controlled trials (RCT), such as “quantity” and “quality” of evidence generated, has lowered healthcare professionals’ confidence in using EBP. An alternate paradigm of Practice-Based Evidence has evolved with the key being evidence drawn from practice settings. Through the use of health information technology, electronic health records (EHR) capture relevant clinical practice “evidence”. A data-driven approach is proposed to capitalize on the benefits of EHR. The issues of data privacy, security and integrity are diminished by an information accountability concept. Data warehouse architecture completes the data-driven approach by integrating health data from multi-source systems, unique within the healthcare environment.

Effective communication during disasters: What effects community decision making & action?

Study/Objective This paper describes a program of research examining emergency messaging during the response and early recovery phases of natural disasters. The objective of this suite of studies is to develop message construction frameworks and channels that maximise community compliance with instructional messaging. The research has adopted a multi-hazard approach and considers the impact of formal emergency messages, as well as informal messages (e.g., social media posts), on community compliance. Background In recent years, media reports have consistently demonstrated highly variable community compliance to instructional messaging during natural disasters. Footage of individuals watching a tsunami approaching from the beach or being over-run by floodwaters are disturbing and indicate the need for a clearer understanding of decision making under stress. This project’s multi-hazard approach considers the time lag between knowledge of the event and desired action, as well as how factors such as message fatigue, message ambiguity, and the interplay of messaging from multiple media sources are likely to play a role in an individual’s compliance with an emergency instruction. Methods To examine effective messaging strategy, we conduct a critical analysis of the literature to develop a framework for community consultation and design experiments to test the potential for compliance improvement. Results Preliminary results indicate that there is, as yet, little published evidence on which to base decisions about emergency instructional messages to threatened communities. Conclusion The research described here will contribute improvements in emergency instructional message compliance by generating an evidence-based framework that takes into account behavioural compliance theory, the psychology of decision making under stress, and multiple channels of communication including social media.

Effective communication during disasters

This paper describes a program of research examining emergency messaging during the response and early recovery phases of natural disasters. The objective of this suite of studies is to develop message construction frameworks and channels that maximise community compliance with instructional messaging. The research has adopted a multi-hazard approach and considers the impact of formal emergency messages, as well as informal messages (e.g., social media posts), on community compliance. In recent years, media reports have consistently demonstrated highly variable community compliance to instructional messaging during natural disasters. Footage of individuals watching a tsunami approaching from the beach or being over-run by floodwaters are disturbing and indicate the need for a clearer understanding of decision making under stress. This project’s multi-hazard approach considers the time lag between knowledge of the event and desired action, as well as how factors such as message fatigue, message ambiguity, and the interplay of messaging from multiple media sources are likely to play a role in an individual’s compliance with an emergency instruction. To examine effective messaging strategy, we conduct a critical analysis of the literature to develop a framework for community consultation and design experiments to test the potential for compliance improvement. Preliminary results indicate that there is, as yet, little published evidence on which to base decisions about emergency instructional messages to threatened communities. The research described here will contribute improvements in emergency instructional message compliance by generating an evidence-based framework that takes into account behavioural compliance theory, the psychology of decision making under stress, and multiple channels of communication including social media.