Research Bank
Not a member yet
5195 research outputs found
Sort by
Taraxacum pseudohamatum Dahlst. (Asteraceae, Cichorioideae): A new naturalisation from the Chatham Islands, Aotearoa / New Zealand
No full textTaraxacum pseudohamatum is reported from Rekohu / Wharekauri / Chatham Island in the Chatham Islands, Aotearoa / New Zealand. This species is an addition to that island group's naturalised flora and the New Zealand Botanical Region. It is also the second member ofTaraxacum Sect Hamatum to be recorded from the islands. The first species of that section to be recorded from the islands is Taraxacu m hamatum, which is still only known from a collection made from Hokorereoro / Rangatira / South East Island. Taraxacum pseudohamatum is locally common on Rekohu / Wharekauri / Chatham Island. The species is described using material from that island, and notes on its recognition and distinction from T hamatum presented The ecology of the species on Rekohu / Wharekauri / Chatham Island is described
Assessment of affective state for horses engaged in therapeutic riding
No full textThe ability to recognise and understand horse behaviour is important to ensure accurate assessments, that can impact equine welfare in New Zealand. This is an important consideration in the management of New Zealand Riding for the Disabled Association’s (NZRDA) horses used for therapeutic riding. A cross-sectional survey was conducted to investigate the ability of NZRDA volunteers, the general public with equine experience, and a panel of equine behaviour experts to identify behaviours and attribute them to an assessment of welfare. Respondents were presented with a selection from 13 videos showing horses engaged in therapeutic riding, and also in a paddock or yard situation, and were asked to select terms that described the affective state of the animal. The videos were scored on a scale ranging from positive to negative affective state. Overall agreement was moderate for videos assessed within the behaviour expert group, but weak between NZRDA volunteers, the general public, and the experts. Agreement was more likely for horses in a paddock/yard context compared to a therapeutic riding context. The low levels of agreement between NZRDA volunteers, the general public, and experts indicate serious concerns for recognising horse behaviours and affective state which, if missed, ignored, or misread could result in serious injury for riders or horses. It is evident that the assessment of equine behaviour as it relates to ‘mood’ is in need of systematic analysis. This would increase the information available to those involved in horse-related activities to be aware of stressors and behavioural indicators that impact horse and rider
RCNN-based analysis of apple trees leaves for early plant disease detection
No full textEarly identification and accurate localization of plant diseases are essential for enhancing crop output and promoting sustainable agricultural practices. This thesis focuses on creating an innovative hybrid deep learning model that combines Region-Based Convolutional Neural Networks (RCNN) and Deformable Convolutional Networks (DCN) to improve the detection and categorization of illnesses in apple leaves. The hybrid technique seeks to tackle difficulties in precisely identifying regions afflicted by illness, particularly when confronted with complex patterns, uneven shapes, and varying sizes of leaf infections. The proposed structure utilizes the advantages of both RCNN and DCN to address the intricacies of disease patterns, including irregular forms, variable sizes, and diverse textures. Region-Based is utilized to provide region suggestions and extract spatial information, facilitating the detection of possibly sick areas within an image. DCN, due to its ability to cope with deformable patterns, improves feature extraction by collecting complex details sometimes overlooked by traditional convolutional networks. This combination guarantees strong efficacy in detecting and pinpointing disease affected regions. A custom dataset was produced for this study, comprising annotated images of apple leaves with clearly defined diseased areas. Preprocessing methods, such as reducing photos to 100×100 pixels and standardizing data formats, were implemented to guarantee compatibility with the model's input specifications. The dataset was utilized to train the hybrid model, which integrates both feature extraction and classification phases. The framework's principal innovations encompass the incorporation of deformable convolutional layers to manage spatial variability and the application of bounding boxes for the localization of impacted regions. The thesis further examines the technical execution of the hybrid RCNN DCN model, clarifying the model architecture, training setups, and preprocessing pipelines. The hybrid model was executed using TensorFlow and trained for 25 epochs with an 80-20 training-testing split. Evaluation metrics, such as Mean Intersection over Union (IoU), Mean Squared Error (MSE), and accuracy, were employed to assess model performance. The experimental results validated the model's efficacy, achieving a mean Intersection over Union (IoU) of 89.57% and an accuracy of 91.48%, markedly surpassing baseline methods. Visual overlays of predictions on test images validated the model's ability to appropriately pinpoint and classify sick areas.
This guarantees scalability and facilitates customisation to suit various datasets and plant types. Employing comprehensive preprocessing and network optimization methods diminishes computing demands while preserving increased accuracy and dependability. This study highlights the capacity of deep learning to transform plant disease detection through automated, scalable, and precise solutions. The hybrid RCNN-DCN framework establishes a basis for future developments in precision agriculture, underscoring its applicability in real-world applications
Healthy lecturers = healthy outcomes
No full textAIMS AND OBJECTIVES
•The research aims to understand the perspectives of staff working in the sector and investigate enablers and barriers to a healthy workplace.
RESEARCH QUESTIONS
1) What are the enablers and barriers to a healthy academic and clinical teaching workplace?
2) How can we support each other in delivering an effective programme to students in class and clinical placement?
3) What are the priorities for the workplace?
4) What must we do now?
5) What must we never do? What is unhelpful?
6) What is the current literature on healthy workplaces for lecturers
Revealing predictive insights: Harnessing statistical and machine learning techniques for New Zealand Stock Index forecasting
No full textThe purpose of this presentation is to evaluate the forecasting effectiveness of the ARIMA model (a statistical approach) in comparison to the LSTM model (an advanced Machine Learning and Deep Learning technique
Simulating consumer behaviour towards hydrogen heavy vehicles
No full textThe growing need to reduce greenhouse gas emissions, particularly in hard-to-abate sectors like heavy transport, has increased interest in alternative fuel technologies such as hydrogen vehicles. Despite their potential, widespread adoption of hydrogen vehicles is hindered by challenges like capital costs. fuel prices and a lack of refuelling infrasture. These barriers critically influence consumer bheaviour and vehicle adoption trends.
This study utilizes the system dynamics model UniSyD_NZ to simulate the adoption of hydrogen and conventional diesel heavy-duty vehicles in New Zealand from 2015 to 2050. Using a nested multinomial logit approach, the model evaluates vehicle uptake across various scenarios. Two distinct utility functions are incorporated to reflect different consumer priorities. The first utility function focuses on fuel availability, addressing concerns about limited hydrogyn refuelling infrastructure. The second utility function considers economic factors, including annual refuelling cost, which is based on four main factors: distance to the nearest station, refuelling and queueing times, and fuel availability risks. Results indicate that hydrogen-diesel dual-fuel vehicles can act as a transitional technolgy, mitigating early adoption barriers where fuel availability if a key concern.
Conversely, in scenarios where fuelling cost dominates consumer decisions, hydrogen vehicle adoption becomes less reliant on dual-fuel vehicles. Nevertheless, the introduction of dual-fuel vehicles significantly reduces the market share of conventional diesel vehicles, contributing to efforts to lower greenhouse gas emissions
Intertwining community-driven and student-built approaches to activate suburban streets
No full textSuburban streets often lack vitality and fail to foster social interaction and community cohesion. Traditional approaches to retrofitting tend to focus on static spatial interventions, proving costly and inadequate for addressing dynamic community needs. This paper explores flexible, cost-effective solutions that can activate suburban streets and enhance cultural cohesion.
Through a brief review of street retrofitting concepts and practices, the authors investigate ways in which an inclusive approach can promote community engagement and transform streets into dynamic social spaces. This approach consists of a community driven and student-built collaborative design, prefabrication and installation, and adaptive use by the community. Examining Tāmaki Makaurau Auckland in Aotearoa New Zealand as a context rich in cultural diversity and with a demand for vibrant public spaces, the authors tested this new approach through design interventions by Unitec architecture students in Open Streets events in Avondale, Tāmaki Makaurau Auckland.
The authors then tested this approach with two projects: Toutai-'a Maui: Maui’s Catch at the Whau Arts Festival, and the Woven Gateway at the We Are Woven Festival. The positive outcomes of these two projects demonstrate the potential of temporary architectural interventions in activating streets and fostering conversation about the streets of Avondale
Value-adding collaborative design and construct sustainability work practices on increasingly complex AEC projects
No full textRESEARCH QUESTION
How to use collaborative design and delivery practices at the preconstruction and construction stages to achieve a circular economy?
ABSTRACT
There is a good amount of research on how collaborative design and delivery practices, such as design & build, integrated project delivery and alliances can help in achieving project goals, sustainability among other things. Although the understanding of these collaborative practices and their advantages has increased, a large part of construction projects are still conventional, design-bid-build projects. The industry is fragmented not only when it comes to separating the design stage from construction, but also when it comes to the number of subcontractors and suppliers in projects. This situation is not going to change any time soon, rather it looks the opposite. Construction projects are becoming more and more complex requiring companies to specify their expertise to a certain design, manufacturing or installation aspect. This is a discovery and theory-led case study that explores key problems and opportunities, and illustrates how theories are applied and adapted, or adopted, in a real-life setting. The focus is on how a top tier main contracting company in New Zealand is changing its practices to be more collaborative by involving the design and construct teams, the client and stakeholders, on a regular workshop basis at the pre-construction design stages in particular. The research question is "How to use collaborative design and delivery practices at the preconstruction and construction stages to achieve a circular economy?" The vision and aim is to explore-with project examples-alternative insightful design strategies, reduce waste to landfill, associated costs, as well as enhancing the environmental impacts on the community. One of the introduced tools so far is an Environment in Design (EiD) register, in conjunction with resource sorter training, and recording all waste data, to achieve the main contractor and key stakeholders' sustainability goals
Post photography and urban skepticism
No full textPost-photographic representations of the cities we inhabit are increasingly exhibiting seamless shifts from the digital to the real, profoundly altering our understanding of time, space, and the urban landscape in ways previously unimaginable. The convergence of digital images, architectural renders, communication networks, and extensive data recording all serve to expand the possibilities for observing and understanding the urban world, allowing for a richer, more comprehensive, and polyphonic account of cities and their complex social dynamics.
However, these new conditions of image production and reception have also brought us to the point of significant contradiction, creating a paradox where we are both more reliant on and more accepting of the digital, even as it contributes to a growing fragmentation of perception. In response to the pervasive seamlessness of digital technologies, some artists have adopted a more sceptical and critical stance. They actively seek out and utilize the fissures, gaps, and glitches within these technologies, which can act to slow down the processes of acceptance and uncover the inherent contradictions, thereby revealing what Franco Berardi (2019) describes as the “inscribed possibilities” within the urban spaces we inhabi
Enhancing students' CV writing skills using Llama 3 large language model
No full textRESEARCH QUESTION
1. Can AI tools effectively help students match their qualifications to specific job requirements through interactive CV writing?
2. Can using the AI tool enhance students' understanding of the basic content of CVs and align their expression of skills with job expectations?
ABSTRACT
In recent years, Artificial Intelligence (AI) technology has had a far-reaching impact on many industries. In the education field, AI is reshaping teaching and learning models to make the learning experience more personalized. And it also has a significant role in improving equity and access to educational resources.
Currently, AI has many applications in education, but there is still a gap in providing support for students preparing for the work. This study focuses on the role of AI in facilitating education, especially in enhancing students' Curriculum Vitae (CV) writing skills. To fill this gap, we have developed an AI-driven CV writing tool that combines Support Vector Machine (SVM) Modeling and Llama 3 Large Language Model (LLM) to provide continuous feedback and improvement to students.
Our approach consists of training an Support Vector Machine model using a dataset, rating CV content by quality, and then using LLM to optimize student responses based on predefined criteria. Iterative rating and cyclic optimization enabled students to refine their CVs until satisfactory results were achieved. In addition, student feedback was collected through a questionnaire that provided insights into the tool's usability, its impact on learning, and the quality of the AI-generated content.
A total of 77 students' feedback on the CV writing program was collected, and they commented positively on the ease of use of the CV and the improvement in CV writing skills. Suggestions for improvement included refining the interface, adding automation features, expanding templates, improving content accuracy, and pointing out strengths and improvement areas.
The findings suggest that an AI-driven tool like the one developed in this study can assist students in learning and applying CV writing skills by providing a structured and interactive learning experience that reduces reliance on templates. This study highlights the potential of AI in personalizing education, making it a reliable tool for job readiness, and the need for further research to optimize the use of AI in education