16168 research outputs found
Sort by
Evaluating riverine flood policy: Land use planning trends in Aotearoa New Zealand
No full textGlobally, the responsibility to develop flood risk policy is often devolved to local government. However, local governments can lack the capacity to keep up with increasing and changing flood risk and information provision without external guidance and support. Central and state governments can deliver policy support and consistency by providing policy direction or standards based on best practice. Due to diverse localised modelling, plans and policies, there is often limited understanding of the nature of flood policy, the degree of variation between localities, and how authorities are improving practice and responding to increasing and changing risk. In this study, we develop and apply an evaluation tool for riverine flood planning that captures the modelling parameters, policies, and information used by regional authorities, distinguishing between traditional and emerging approaches. We examine three primary categories of regional flood policy: modelling parameters and associated planning regulations, risk-based policy approaches, and information provision processes. Our findings reveal evolving practices, policy variances, and aspects of contention, demonstrating where central and state governments can provide greater direction for policy development. Our evaluation tool therefore provides a basis to guide complex policy transitions, from static hazard-based planning towards a more comprehensive, risk-based approach
Can't we talk about class?: Aotearoa-New Zealand's "classless" discourse
No full textIn Aotearoa-New Zealand (ANZ), class discourse is notably absent from mainstream socio-cultural-political narratives. The perception of an egalitarian society is challenged by its continuous promotion of neoliberal ideology, policy and practice, resulting in a fractured society. As a colonised nation, this has exacerbated poor class relations and limited social mobility. This paper examines ANZ's complex political history to better understand how working- class individuals and communities are often marginalised or made invisible within cultural, political, and social frameworks. By analysing these issues, this paper aims to lay the foundation for a broader conversation within and about ANZ's working-class communities
Motivation factors for students using Generative AI
No full textGenerative Artificial Intelligence (GenAI) has caused a shift in approaches to assessment and academic integrity in tertiary institutions. Recent research internationally and within Australasia underscores the need for responsible GenAI conduct and preparation for a future where work and education are shaped by GenAI technologies. However, ethical considerations often need to be made based on the impact on human research, intellectual property rights, and AI literacy in higher education. Academic integrity and ethical considerations should be used to balance hasty approaches to GenAI to ensure tertiary institutions provide inclusive learning opportunities. All of these impact on the need to understand what motivates student use of GenAI. Using the framework proposed by Bouteraa et al. (2024) as a model, this presentation applies data and results from recent literature to explore the factors which motivate use of Generative AI in a tertiary education context. These factors consist of Performance Expectancy, Effort Expectancy, Technological Self-Efficacy, Educational Self-Efficacy, Integrity, and Personal Anxiety. The presentation highlights what factors make students likely to use Generative AI, what factors demotivate use, and how motivation and demotivation do not always produce a desired outcome
Piri Pāua: Mātauranga Māori and marine science approach to growth rate and length at maturity of pāua in the Bay of Plenty 2022-2024
No full textPāua is a highly regarded taonga (culturally important) species. Mai i ngā Kurī ā Whārei ki Tihirau (the Bay of Plenty Regional Iwi Customary Fisheries Forum) raised concerns about the current state of pāua in Te Moana-a-Toi (Bay of Plenty). This project utilised localised intergenerational observations of population dynamics in traditional harvesting areas to assess pāua productivity in the Tauranga Moana Mātaitai Reserve in Tauranga and Te Rohe Moana o Ngāti Awa in Whakatāne. The project combined mātauranga Māori alongside marine science field methods.
Results were compared with previous iwi-led research at Tauranga in 2013 and Whakatāne in 2010. The pāua population in 2023 had declined by almost half in Tauranga but remained relatively consistent across all sites in Whakatāne. However, pāua were small sized in both locations, with less than 1% of individuals reaching the Minimum Legal Size (125 mm) for harvesting. Growth rate surveys were conducted in the wild at both locations from June 2023 to the end of May 2024 and identified pāua as slow growing and, as a result, sexually mature at smaller sizes than in other regions of Aotearoa New Zealand
A multi-phase assessment for selecting an augmentative and alternative communication modality
No full textChildren with autism, who have limited speech, are often candidates for augmentative and alternative communication (AAC) modalities to learn basic mands. However, few studies have evaluated the assessment of AAC modalities. We report on the results of an evidenced-based multi-phase assessment, with a focus on choice as a foundational element, to evaluate modality selection, comparison, and acquisition for six children with autism. Assessment procedures involved using an indirect assessment that evaluated environments and the caregiver’s preference as a listener. The results of the indirect assessment informed the experimental evaluation of learner acquisition and preference for a modality. Findings indicate that the assessment process is relatively quick, the child participants did demonstrate a preference for a mand modality, and the child participants were able to meet mastery criteria for the use of the initial mand. Results point to a potentially useful approach for assessing AAC modalities for young children with autism
Linguistics of social media: An introduction to the special issue
No full textIt is hard to remember a world without social media (SM). The Internet Relay Chat (IRC) and bulletin boards of the 1980s and the popular early platforms of the 1990s, MySpace, AOL Messenger, and Facebook, gave rise to nothing short of an SM empire in just 50 years. Although we have been considering how language is used on the Internet for some time now, studying the language used on SM is comparatively a newer field of inquiry
Wellbeing and policy in New Zealand: From a wellbeing framework to a government-wide approach
No full textThis chapter summarises the modern state of wellbeing and public policy in New Zealand. The evolution of New Zealand’s wellbeing approach to policy is explained, key learnings from the process are discussed, and specific action points to further wellbeing and public policy in New Zealand and around the world are suggested. New Zealand often ranks in or near the top 10 nations on international wellbeing metrics and is often looked to as one of the leaders in wellbeing policy. In 2011, the New Zealand Treasury reinterpreted its goal to that of promoting the wellbeing of New Zealanders - publishing the Living Standards Framework, a new model of wellbeing for policy-making. Progress on this was greatly aided in 2018 when a change in government brought with it an explicit wellbeing approach to the national budget and reporting processes, much of which was codified into new laws. The Living Standards Framework and its associated wellbeing tools have continued to evolve. Key innovations include an interactive dashboard of wellbeing data and a wellbeing Cost Benefit Analysis tool that covers many domains and indicators of wellbeing. This has resulted in greater collaboration between government agencies and more transparent reporting on the inner workings and results of public policies. But more could be done, including taking environmental concerns more seriously, fixing data gaps, reporting on wellbeing data more frequently, further training and teamwork for policy-makers, and setting up a citizens assembly on wellbeing and public policy
Validating federated learning performance in practice: An agricultural edge hardware testbed analysis
No full textSmart agriculture, driven by the Internet of Things (IoT) and artificial intelligence, generates vast amounts of data from distributed sensors and devices on farms. Traditional centralised machine learning approaches struggle in rural settings due to intermittent connectivity, limited bandwidth, and data privacy concerns. Federated learning (FL) has emerged as a promising approach to address these challenges by collaboratively training models directly on edge devices, for example farm sensors and edge computers, without sending raw data off-site. This thesis presents the design, development, and evaluation of a standards-driven, hardware-based federated learning testbed for smart agriculture. The testbed consists of six NVIDIA Jetson Nano edge computing nodes. A primary objective was to evaluate and compare modern, open-source FL frameworks in a realistic edge environment. Therefore, we specifically tested the FLIGHT framework. FLIGHT was selected for its notable features, including a Function-as-a-Service (FaaS) architecture facilitating serverless deployment and native support for hierarchical topologies relevant to distributed farm networks. We developed a reproducible methodology for deploying and benchmarking FLIGHT on these resource-constrained devices, using the Fashion-MNIST image classification dataset as a proxy for agricultural sensor data to compare performance against known benchmarks and simulation expectations. Key contributions include: 1) the physical testbed itself; 2) an open and repeatable experimental framework centered around FLIGHT; and 3) a comprehensive performance analysis under realistic network conditions and device constraints, providing data to bridge the gap between simulation results and practical deployment challenges. The results demonstrate that the federated model achieves competitive accuracy while significantly reducing raw data transfer. Findings indicate that careful configuration of FL can mitigate the impact of limited connectivity, and that even low-power devices can collaboratively train useful models within reasonable time-frames. These insights validate the viability of FL in smart farming scenarios. The developed testbed and its accompanying benchmarking methodology lay a foundation for future research and deployment of FL in agriculture, bridging the gap between theoretical simulations and real-world farm deployments. This work’s significance lies in providing both a practical tool for researchers to rigorously evaluate FL strategies in edge environments and guidance for stakeholders aiming to deploy privacy-preserving AI in agriculture at scale
Improving finite sample estimates of principal components
No full textPrincipal Component Analysis (PCA) is a method of compressing high-dimensional data into a lower-dimensional format that captures the essence of the original structure. PCA is a matrix decomposition technique based on eigen decomposition. It quantifies relationships between variables using covariance matrices, captures the shape of the data distribution, and evaluates the importance of directions using eigenvalues. Therefore, the accuracy of the variance-covariance estimation is crucial for reliable PCA. In high-dimensional settings where the number of observations (n) is much smaller than the number of variables (p) (i.e., n << p), the conventional Maximum Likelihood Estimator (MLE) of covariance becomes poorly conditioned and yields unreliable principal components. To address these limitations, we propose a novel estimation framework called Pairwise Differences Covariance (PDC), along with four regularized extensions: Standardized PDC (SPDC), Local Scaled PDC (LSPDC), Maximum Absolute Scaled PDC (MAXPDC), and Range
Scaled PDC (RPDC). These estimators increase the effective sample size by utilizing all pairwise differences within the data, thereby enhancing estimation stability without requiring additional data collection. Extensive experiments on synthetic and real datasets demonstrate that the proposed
estimators, particularly SPDC, significantly reduce the over-dispersion of the first principal component and improve directional accuracy. On average, SPDC reduced cosine similarity error by approximately 10–30% and narrowed eigenvalue spread by 10–20% compared to MLE and Ledoit-Wolf estimators in n << p HDLSS scenarios. Real-world applications confirm the practical utility and robustness of these methods for analyzing high-dimensional data
Early career academics navigating the ecology of the university: a collaborative autoethnography
No full textEarly career academics (ECAs) are negatively affected by the neoliberal university which encourages performativity, competition and a ‘publish or perish’ mentality. In this paper a group of four ECAs in the Aotearoa New Zealand context explore and navigate the neoliberal university through a collaborative autoethnography. Collectively, we adopt Barnett’s [2018. The ecological university: A feasible utopia. Routledge, Taylor & Francis Group.] five dimensions of ecologies framework to study our own experiences. Findings suggest that as ECAs, we seek connection to academia, experience workload and power imbalance, and actively learn how to navigate roles. We also describe how our research group created a supportive environment within the competitive space of academia that fostered feelings of belonging and offered support in navigating the university ecology. Furthermore, working in initial teacher education, as a high service discipline, created unique challenges for us such as high service and teaching roles which may negatively impact ECAs. Through this lens of initial teacher education, we in turn make recommendations for universities to better support early career academics