Monash University Research Portal
Not a member yet
41460 research outputs found
Sort by
Automatic Short Answer Grading in the LLM Era:Does GPT-4 with Prompt Engineering beat Traditional Models?
No full textAssessing short answers in educational settings is challenging due to the need for scalability and accuracy, which led to the field of Automatic Short Answer Grading (ASAG). Traditional machine learning models, such as ensemble and embeddings, have been widely researched in ASAG, but they often suffer from generalizability issues. Recently, Large Language Models (LLMs) emerged as an alternative to optimize ASAG systems. However, previous research has failed to present a comprehensive analysis of LLMs' performance powered by prompt engineering strategies and compare its capabilities to traditional models. This study presents a comparative analysis between traditional machine learning models and GPT-4 in the context of ASAG. We investigated the effectiveness of different models and text representation techniques and explored prompt engineering strategies for LLMs. The results indicate that traditional machine learning models outperform LLMs. However, GPT-4 showed promising capabilities, especially when configured with optimized prompt components, such as few-shot examples and clear instructions. This study contributes to the literature by providing a detailed evaluation of LLM performance compared to traditional machine learning models in a multilingual ASAG context, offering insights for developing more efficient automatic grading systems.</p
The Company You Keep:Refining Neural Epistemic Network Analysis
No full textCollaborative problem-solving (CPS) is defined as an inherently sociocognitive phenomena. Despite this, extant learning analytic techniques tend to focus on either the social or cognitive aspects without explicitly considering their interaction. Prior work developed Neural Epistemic Network Analysis (NENA), which used a combination of deep learning methods to simultaneously model the social and cognitive aspects of CPS; however, the method had several limitations. The refined version of NENA presented here addresses these limitations by (a) introducing a simplified autoencoder deep learning architecture; (b) using a combination of social and epistemic networks as input to preserve interpretability in terms of social and cognitive factors; and (c) introducing an isometry loss function to ensure downstream statistical tests are meaningful. We found that the refined version of NENA is able to achieve high performance on criteria we would expect from a network analytic technique in the context of learning analytics: interpretability, goodness of fit, orthogonality and isometry; and discriminatory power. We also demonstrated that this method was comparable in performance to a more traditional learning analytic technique, Epistemic Network Analysis (ENA), while providing information that ENA did not. The results suggest that NENA could be a useful method for exploring the cognitive interactions of a given individual's social network and thus the influences their network exerts upon them.</p
Women's authorship in international human resource management research:implications for responsible management education and emerging scholars
No full textLike many professional occupations, the participation of female scholars has steadily increased since the 1990s in the human resource management (HRM) fields. While it is widely acknowledged that workforce diversity brings different perspectives, we lack insight into the impact of such changes. In this paper, we explore the implications of gender in the authorship of scholarly articles for the knowledge base of this field, using an example of a content analysis of 890 articles in the international HRM field. We discuss the implications of gender in scholarly work both within and beyond the HRM field. We draw connections to the sustainability development agenda and responsible management education from a gender perspective and offer suggestions for the career development of emerging and future scholars.</p
Why do ‘so’ much behind the wheel? How obsessive-compulsive symptoms, mindfulness, and anxiety influence distracted driving behaviours
No full textDriver Distraction Behaviour (DDB) contributes significantly to motor vehicle crashes, psychological factors playing a crucial role in its occurrence. This study aimed to examine the impact of Obsessive–Compulsive Symptoms (OCS), mindfulness, and driving-related anxiety on DDB, as well as the roles of Perceived Safety (PS), Perceived Risk (PR), and Perceived Behavioural Control (PBC). A total of 539 participants (53.6 % female; mean age = 39.6 years, SD = 8.5, range = 21–66 years) completed an online survey assessing self-reported DDB and the six aforementioned factors. Structural Equation Modelling (SEM) was used to examine the relationships among psychological factors, while one-way ANOVAs assessed the effects of individual characteristics on psychological variables. As expected, the SEM results revealed significant positive correlations between OCS, driving-related anxiety, PS, PBC, and DDB, while mindfulness showed a significant negative correlation with DDB. PR had no significant association with DDB. These findings suggest that mindfulness practices may help reduce DDB, whereas anxiety may exacerbate it. OCS, a prevalent negative psychological symptom, may impair mindfulness and, in turn, increase DDB. One-way ANOVA results showed that gender, age, education level, and driving characteristics (i.e., driving age, driving frequency, and annual mileage) significantly influenced some latent variables. With the fast pace of modern life, more individuals engage in non-driving-related tasks while driving. Therefore, further research is needed to explore how mindfulness interventions and strategies to alleviate OCS and anxiety can mitigate DDB and reduce traffic risks. Additionally, traffic safety agencies should implement targeted education programs to address drivers’ overestimation of their control over DDB, reinforce awareness of its risks, and ultimately decrease its prevalence.</p
A geochemical survey of the Antas Valley, Sardinia:Medieval metallurgy and modern slag recycling?
No full textGeochemical surveys of the Antas Valley in Sardinia, Italy, have revealed significant zinc and lead concentrations along the Antas River floodplains, suggesting the presence of medieval ore-processing workshops which are otherwise hard to detect. While the zinc concentrations were found to be dispersed and probably related to the erosion of zinc-rich dolomite, the lead concentrations were more localised on the banks of the river, suggesting an anthropogenic origin. Three large concentrations of lead were found to coincide with deposits of black glassy slag, a by-product of ore processing. Analysis of the slag revealed a high lead content (around 75%) and very low zinc content, which along with historical research suggests either medieval ore processing for silver and/or 19th-century slag recycling. The near absence of zinc in the slag supports the 19th-century recycling hypothesis, as this period saw an increased demand for silver, lead and zinc and the development of processes to extract it from older slag. Further research, including dating of the slag and excavation of the lead-enriched areas, is required to confirm the origin of the slag deposits.</p
Five Top Issues for Young Voters in The 2025 Federal Election:Insights from The Australian Youth Barometer
No full textDrawing from CYPEP’s Australian Youth Barometer and other recent surveys of young people, this report summarises the key issues identified by young Australians that we believe should inform political parties’ policy platforms in the run-up to the 2025 Australian federal election. Numerous surveys in Australia have identified a variety of issues that overlap in significant ways. For example, The Salvation Army’s 2021 Social Justice Stocktake asked Australians to reflect on social justice in their communities. The top five issues Australians wanted addressed were mental health (53.9%), housing affordability (52.4%), alcohol and drug misuse (42.6%), family violence (35.4%), and homelessness (35.1%). According to Mission Australia, young people aged 15–19 identified the most important issues in Australia today as the cost of living, climate change and the environment, violence, safety and crime, and mental health. Our own survey of young people aged 18–24, the 2024 Australian Youth Barometer, found the top five issues that young people think needed immediate action in Australia were affordable housing options for young people (73%), employment opportunities for young people (52%), climate change (40%), race relations and racial inequality (32%), and gender inequality at work and in public places (29%).This paper focuses on the following five issues identified by young people across several surveys:1. Housing affordability2. Employment and finances3. Climate change4. Inequality and discrimination5. Health and mental health∗In our discussion of each issue, we highlight the significant findings that should inform any policy responses
Predict+Optimize Problem in Renewable Energy Scheduling
No full textPredict+Optimize frameworks integrate forecasting and optimization to address real-world challenges such as renewable energy scheduling, where variability and uncertainty are critical factors. This paper benchmarks solutions from the IEEE-CIS Technical Challenge on Predict+Optimize for Renewable Energy Scheduling, focusing on forecasting renewable production and demand and optimizing energy cost. The competition attracted 49 participants in total. The top-ranked method employed stochastic optimization using LightGBM ensembles, and achieved at least a 2% reduction in energy costs compared to deterministic approaches, demonstrating that the most accurate point forecast does not necessarily guarantee the best performance in downstream optimization. The published data and problem setting establish a benchmark for further research into integrated forecasting-optimization methods for energy systems, highlighting the importance of considering forecast uncertainty in optimization models to achieve cost-effective and reliable energy management. The novelty of this work lies in its comprehensive evaluation of Predict+Optimize methodologies applied to a real-world renewable energy scheduling problem, providing insights into the scalability, generalizability, and effectiveness of the proposed solutions. Potential applications extend beyond energy systems to any domain requiring integrated forecasting and optimization, such as supply chain management, transportation planning, and financial portfolio optimization.</p
Exploring the effect of driver drowsiness on takeover performance during automated driving:an updated literature review
No full textIntroduction: Vehicle automation technology has considerable potential for reducing road crashes associated with human error, including issues related to driver drowsiness. However, before full automation becomes available on public roads, it will be essential for drivers to take back control from automated driving systems when requested. This poses a challenge for drivers, particularly as automation may further exacerbate drowsiness. This paper aims to update a systematic review published in 2022 (Merlhiot & Bueno, Accident Analysis and Prevention, 170, 106536), to discuss factors affecting driving drowsiness and takeover performance with a particular focus on those not identified in previous review. Method: Following the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines, three databases: Web of Science, PubMed and Scopus were searched for studies published between March 2021 and October 2024. The following eligibility criteria were applied for study inclusion: 1) participants must have interacted with a simulated or real-world vehicle featured with driving automation Level 2 or above; 2) with at least one measurement indicator of driver drowsiness; 3) with at least one measurement indicator of takeover performance; 4) be conducted within a controlled experimental design. From an initial selection of 182 articles from databases, a total of twelve published articles were obtained after removing duplicates, title, abstracts and full texts checking. Additionally, 17 articles from the previous review were included, resulting in a total of 29 articles for this review study. Results: Driver drowsiness (e.g, increased Karolinska Sleepiness Scale levels, blink frequency) tended to increase with both the duration of automated driving and automation levels. Engaging in non-driving related tasks (NDRTs) alleviates drowsiness (e.g, lower heart rate and percentage of eye closure), but reduces takeover performance (e.g., longer braking reaction times, stronger longitudinal acceleration, shorter minimal time to collision). Compared to older drivers, younger drivers were more susceptible to drowsiness, while older drivers had worse takeover performance (e.g., delayed steering reaction time, higher collision rates). Sleep inertia and circadian rhythms were also identified as factors influencing takeover performance. The road monitoring task helps prevent excessive participation in NDRTs and improves takeover performance (e.g, reduced brake reaction times and maximum steering velocity, increased the minimum time to collision). Digital voice assistants and scheduled manual driving help maintain alertness (e.g, decreased blink duration) and enhance takeover performance (e.g, shorter reaction time to resume steering). There were several limitations of the methodologies applied in the existing studies, among which were: 1) a lack of verification through real-world driving experiments; 2) insufficient diversity in the measurement of driver drowsiness; 3) singularity of takeover scenarios; 4) failure to reveal the mechanism by which drowsiness affects takeover performance. Conclusion: Factors such as duration of automated driving, NDRT engagement, driver age, sleep-related issues and automation levels influence the development of drowsiness and subsequent takeover performance. This literature review highlights several necessary directions for future research: 1) what underlying factors affect drowsiness and take over performance; 2) how to prevent the occurrence of driver drowsiness; 3) how to alleviate driver drowsiness once it occurs; 4) how to assist drowsy drivers to regain control of the vehicle safely and quickly.</p
Managing cyber harm:a survey of challenges, practices, and opportunities
No full textCurrent approaches in cybersecurity predominantly focus on identifying threats and vulnerabilities, often overlooking the qualitative and cascading aspects of harm from cyber incidents. However, the increasing dependency on internet-connected devices and the rapid evolution of digital technologies have heightened the potential for cyber harm. As a result, there is a growing recognition of the importance of focusing on harm to improve cyber protection and response capabilities. Despite this shift in focus, a comprehensive review of existing literature focusing on cyber harm remains absent. Such a review is crucial to understanding the intricacies of cyber harm and developing appropriate management strategies. This paper aims to address this gap by presenting an extensive review of recent literature on cyber harm, examining diverse types of harm arising from cyber incidents, and identifying key challenges in managing them. In addition, national and international efforts to prevent and mitigate cyber harm are also discussed. This work significantly advances a holistic understanding of cyber harm, a necessary step toward developing a Cyber Harm Model (CHM) that will enable researchers and practitioners to assess and mitigate the complex consequences of cyber incidents. The analysis has theoretical and policy implications, enhancing academic knowledge and informing national and international cybersecurity efforts.</p
University student disclosures of crime, violence, and trauma:findings from a survey of criminology educators across Australia and Aotearoa New Zealand
No full textThis study explores the findings from a survey-based questionnaire investigating the prevalence and predictors of student disclosures of crime, violence, and trauma to criminology educators working at Australian and Aotearoa New Zealand universities. Responses show student disclosures are common, with educators receiving an average of three to four disclosures in the preceding two years. While gender did not predict the number of disclosures received, teaching subjects discussing domestic and family/whānau and/or sexual violence increased the likelihood of disclosures. The study’s findings can help inform the development of university interventions, systems, and resources to improve support for students and staff, enhancing classroom and campus safety.</p