Research Bank
Not a member yet
5195 research outputs found
Sort by
Puna Kōrero Project: [United Matariki celebration]
No full textLaunch of Allana Goldsmith’s new Puoro Ataata (Music Video)
Puna Kōrero Drawing and Animation Project
Campus Storytelling
"Unitec Matariki 2024"
Video of highlights from celebration:
https://vimeo.com/97784686
Optimizing plant disease classification with hybrid convolutional neural network–recurrent neural network and liquid time-constant network
No full textThis paper addresses the practical challenge of detecting tomato plant diseases using a hybrid lightweight model that combines a Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN). Traditional image classification models demand substantial computational resources, limiting their practicality. This study aimed to develop a model that can be easily implemented on low-cost IoT devices while maintaining high accuracy with real-world images. The methodology leverages a CNN for extracting high-level image features and an RNN for capturing temporal relationships, thereby enhancing model performance. The proposed model incorporates a Closed-form Continuous-time Neural Network, a lightweight variant of liquid time-constant net works, and integrates Neural Circuit Policy to capture long-term dependencies in image patterns, reducing overfitting. Augmentation techniques such as random rotation and brightness adjustments were applied to the training data to improve generalization. The results demonstrate that the hybrid models outperform their single pre-trained CNN counterparts in both accuracy and computational cost, achieving a 97.15% accuracy on the test set with the proposed model, compared to around 94% for state-of-the-art pre-trained models. This study provides evidence of the effectiveness of hybrid CNN-RNN models in improving accuracy without increasing computational cost and highlights the potential of liquid neural networks in such applications
Ordered phases in ternary wurtzite group-III nitrides: A first-principles study
No full textA first-principles-based lattice model is applied to investigate the ordered phases of mixed group-III nitride ternary alloys. The model surveys the atomistic configurations with the lowest formation enthalpy for a wide range of compositions. We found novel ordered phases in wurtzite structures having specific compositions of three- and four-sevenths molar fractions of group-III cations. The configurations of group–III atoms on cation sites in those phases consist of a characteristic fragment of the ordered phases of one-third and one-half ordered phases that were previously reported. The simulation results indicate that group-III cations in ternary nitrides follow spatial positioning “atomistic distancing rules” that can be described by the pairwise interaction energy of group-III cations to realize the stabilities of the ordered structures. To minimize the formation enthalpy of a mixed crystal, the minor B, Al, Ga, and In atoms on cation sites remain neither too close to nor too distant from each other, allowing those ordered phases to be realized
Design for wellbeing: Enhancing wellbeing and quality of life in Avondale
No full textARCHITECTURAL QUESTION
How can a high-density, medium-rise housing design enhance wellbeing and quality of life in the Auckland suburban area of Avondale?
ABSTRACT
Designing the urban and built environment to enhance physical conditions has long been a concept in planning and design. However, the recent emergence of the COVID-19 pandemic has become a catalyst for realizing and appreciating the importance of mental health in the process of designing for wellbeing.
Avondale, in Tamaki Makaurau Auckland, is a culturally diverse and growing suburb that requires increased housing density. The growing New Zealand population, particularly in Tamaki Makaurau Auckland, has made intensification a pressing issue. An expanded focus towards housing models has shifted from standalone homes to larger scale solutions, such as higher-density housing of a medium-rise nature. This type of housing design can enhance wellbeing as cities continue to grow and intensify in the urban occupation.
The COVID-19 pandemic changed how people interacted with each other, with lockdowns and social distancing negatively impacting people’s wellbeing, and highlighting the importance of community interaction. Approaches to designing urban and built spaces that promote mental health are outlined in the text Restorative Cities: Urban Design for Mental Health and Wellbeing.
Key concepts for designing urban spaces for mental health suggest placing greater priority on the interaction between place and health. By addressing traditional barriers, urban and built design has great potential to create restorative cities that correlate with increased mental health and wellbeing. The environment should be designed according to the core principles of walkability, social interaction, activity, and life, with spaces woven into the physical and social fabric of society.
This project will draw on existing knowledge of how to design spaces for enhanced wellbeing and quality of life, building on architectural and other theories to improve perceived experiences of the built and urban environment. The research will provide an approach to promoting mental health and wellbeing in the design of future dwellings. Wellbeing is crucial to how people live, grow, and perceive themselves and others.
The resulting project framework seeks to create a diverse housing model that caters to all demographics. The housing model will implement communal facilities and a range of residential typologies. The project will also incorporate a scheme with retail, eateries, and community-related spaces to rejuvenate the outdated Avondale commercial district. Overall, the project aims to reflect the suburb's diverse community and create spaces that enable the community to thrive amid urban intensification
A comparative analysis of machine learning algorithms in network-based intrusion detection systems for detecting advanced persistent threats to enhance cybersecurity
No full textRESEARCH QUESTIONS
Q1: Which ML algorithms most effectively detect APT traffic within NIDS?
Q2: What network behaviour features are most significant for identifying APTs using ML models?
Q3: How does data imbalance impact the detection accuracy of ML algorithms, and how do resampling techniques impact their performance?
ABSTRACT
Advanced Persistent Threats (APTs) pose a sophisticated and evolving cyber risk, persistently targeting organisations to extract high-value information. Traditional Intrusion Detection Systems (IDS), which rely on signature-based and heuristic methods, often fail to detect APTs due to their stealthy nature, high false positive rates, and inability to generalise across diverse attack patterns. Additionally, prior research on ML-based IDS has been limited by challenges such as data imbalance, ineffective feature selection strategies, and inconsistent model performance across different datasets.
This research addresses these gaps by performing a comparative evaluation of classical ML and deep learning algorithms for APT detection using the NF-UQ-NIDS-v2 dataset. This large and diverse NetFlow-based dataset is the combination of 4 datasets such as NF-UNSW NB15-v2, NF-BoT-IoT-v2, NF-ToN-IoT-v2, and NF-CSE-CIC-IDS2018-v2) comprising 43 features and over 75 million records, including 21 attack types. A robust ensemble-based feature selection approach was implemented to identify the most relevant network traffic features, optimising model performance. Nine classical ML models, including CatBoost, Decision Tree, K-Nearest Neighbors, Gradient Boosting, Random Forest, SVM, XGBoost, and two deep learning models, 1D-CNN and LSTM, were trained and evaluated using a K-fold cross-validation strategy to ensure reliability.
The results demonstrate that ensemble learning models outperformed other approaches in detecting APTs. For multi-class classification, XGBoost achieved the highest accuracy (93.23%), followed by Gradient Boosting (92.75%) and Random Forest (92.50%). In binary classification, XGBoost and Random Forest achieved outstanding accuracies of 96.40% and 96.37%, respectively, while CatBoost closely followed at 96.13%. These findings highlight the effectiveness of tree-based ensemble methods in capturing complex attack patterns and reducing misclassification rates, making them highly suitable for modern IDS. This study contributes to the field by presenting a comprehensive evaluation of ML models for APT detection, refining feature selection techniques, and providing insights into the strengths and limitations of various approaches for real-world IDS deployment
A journey through the senses: An approach towards Hamilton’s multi-sensory transport hub
No full textRESEARCH QUESTION
How can multi-sensory architecture provide an effective and enriching experience for all users?
ABSTRACT
Sensory perception is the cornerstone of perceiving architecture and the surrounding environment. Modern architecture is primarily designed for the visual sense, which overrides the other four senses and limits sensory experiences for most people. The visual bias in architecture raises concerns about the sense of inclusiveness. The notion of empathy bridges the emotional gap between low vision and visual-centric architecture. The distinctive spatial perception and heightened senses of people that have low vision provides a unique opportunity to enhance experiences through multiple senses. The approach to multi-sensory architecture ensures the privileged visual sense is overturned to reinforce bodily connections through multi-sensory experiences that are effective and enriching for all users.
How can multi-sensory architecture provide an effective and enriching experience for all users?
The design of a transport hub will form the basis of addressing this question. Transport hubs play a significant role in many individuals’ unique day to day journeys. Multi-sensory architecture aims to engage and stimulate all five senses beyond just the visual sense. Incorporating all five senses within a transport hub seeks to provide effective and enriching multi-sensory experiences for all through the exploration and integration of visual, auditory, tactile, olfactory and gustatory elements, reinforced by aspects of materiality, movement and biophilia.
Enhancing the use of public transport can have a positive impact on society as it is beneficial for everyone. The future of railway transport envisions a well connected inter-regional connection throughout the North Island, making Hamilton the nearest major station to Auckland’s transport network. An existing abandoned railway station located in Hamilton Central will be the catalyst of this project, which aims to provide enriching sensory experiences for everyone
Reaching across difference
No full textThere are growing numbers of trans and non-binary whānau accessing perinatal healthcare in Aotearoa New Zealand (NZ). International literature shows that their perinatal healthcare needs, including their lactation and chestfeeding/breastfeeding information, care and support needs, are not well understood by perinatal healthcare providers. This leads to inequities in perinatal care provision and access, potentially jeopardizing reproductive justice for trans and non-binary whānau. My study aimed to explore the knowledge and beliefs of Lead Maternity Care (LMC) kahu pōkai about lactation and chestfeeding/breastfeeding for trans and non-binary whānau with the purpose of understanding continuing midwifery education needs in this space.
A qualitative interpretive description study design, underpinned by social constructionism, facilitated the identification of potential causes of health inequity while recognising the social, cultural and historical contexts within which knowledge is constructed, and the power dynamics that influence this construction. The ability to determine what may be required to interrupt potential reasons for inequity is enhanced by this understanding.
I conducted semi-structured interviews with ten LMC kahu pōkai. I analysed the data garnered from these interviews using reflexive thematic analysis and identified four themes. The first theme illustrated participants’ awareness of their lack of knowledge and explored why and how they tried to address their knowledge gaps. Theme two described participants’ perception of a system-wide lack of knowledge due to cis-normative and gender essentialist assumptions and depicted participants’ sense of responsibility to keep whānau safe within this system. Theme three explored what participants were doing to advocate for trans and non-binary whānau within their sphere of influence and identified what support is needed from the midwifery profession to generate a broader impact. And finally, the fourth theme provided participants’ perspectives for an expansive imagining of midwifery that safely serves all whānau who access perinatal healthcare.
Findings demonstrate that participants are motivated to learn and desire education that prepares them to provide safe and inclusive perinatal care for trans and non-binary whānau generally before focusing on the specificity of lactation and chestfeeding/breastfeeding care. For whānau to have the best chance of experiencing reproductive justice from midwifery care in Aotearoa NZ, profession-wide education and engagement with cultural humility is required to address and unpack the restrictive cis-normative and gender essentialist assumptions that have become embedded norms within the midwifery profession in Aotearoa NZ
The Womb Chair Speaks
No full textAs part of International Women’s Day, 8 March, the University of Auckland Creative Arts and Industries, School of Architecture and Planning hosted a talk on the Womb Chair, designed by renowned mid-century modernist architect Eero Saarinen. Initially commissioned by Florence Knoll in 1948, Saarinen’s Womb Chair challenged the conventions of chair design of its time. It pushed the boundaries with its innovative use of technologies: a departure from the traditional techniques of the era. This allowed for a form and functionality that deviated from traditional prescriptive norms, marking a significant shift in design philosophy
The Mind and Body Centre: A go-to place for growth and comfort
No full textRESEARCH QUESTION
How can the integration of architecture and nature be optimized to positively impact mental health and well-being?
ABSTRACT
The COVID-19 pandemic has brought to light many nations' unpreparedness to address a surge in mental health challenges. This crisis has exposed systemic issues like limited treatment options, a scarcity of healthcare professionals, and inadequate access to natural environments, which can help alleviate distress during lockdowns. Mental health facilities and professionals were overwhelmed by the influx of patients, resulting in severe consequences for those in need.
The surge in mental health issues during lockdown sparked concerns about a sudden onset of unpleasant feelings. Widespread sentiments of loneliness exacerbated many people's health concerns. Though this period eventually ended, and life returned to normal, the experience had a long-term influence. It serves as a reminder that some people relied only on themselves throughout the pandemic for survival. This emphasizes the importance of evaluating how our built environment contributes to resilience and coping approaches during crises like this.
This project aims to redefine how mental healthcare facilities tackle mental health concerns, striving to eradicate associated stigmas. This study will explore the concept of a sanctuary—a welcoming community hub doubling as a healthcare facility. It will address and alleviate mental health issues and prioritize individuals' overall well-being. The facility aims to bolster resilience by fostering connections and nurturing personal growth, empowering individuals to navigate potential future pandemics and their daily challenges effectively.
Mental healthcare facilities often overlook the psychological benefits of having access to and connection with nature. This oversight has a substantial impact on how individuals perceive mental health and their willingness to seek access to services. This project is situated within Newmarket's metropolitan cityscape, devoid of green spaces and dominated by filled parking lots and bustling streets. As a result, the study will look thoroughly into the use of biophilic, salutogenic, and restorative strategies to improve the design of mental health facilities. The goal of revitalizing the site through design is to create a supportive retreat from life's pressures, aid in the recovery from mental health issues, and promote a positive approach to managing work-life balance.
The research outlined in this thesis highlights deficiencies in mental health services and the number of people affected. It also explores how architecture and nature can transform a place into a mental healthcare facility that cares for the ill and nurtures and maintains healthy individuals' well-being
AI-enhanced personality identification of websites
No full textThis paper addresses the challenge of objectively determining a website’s personality by developing a methodology based on automated quantitative analysis, thus avoiding the biases inherent in human surveys. Utilizing a database of 3000 websites, data extraction tools gather relevant data, which are then analyzed using Artificial Intelligence (AI) techniques, including machine learning (ML) and natural language processing. Four ML algorithms—K-means, Expectation Maximization, Hierarchical Agglomerative Clustering, and DBSCAN—are implemented to assess and classify website personality traits. Each algorithm’s strengths and weaknesses are evaluated in terms of data organization, cluster flexibility, and handling of outliers. A software tool is developed to facilitate the research process, from database creation and data extraction to ML application and results analysis. Experimental validation, conducted with identical training and testing datasets, achieves a success rate of up to 94% (with an Error of ≤50% ) in accurately identifying website personality, which is validated by subsequent surveys. The research highlights significant relationships between website attributes and personality traits, offering practical applications for website developers. For instance, developers can use these insights to design websites that align with business goals, enhance customer engagement, and foster brand loyalty. Additionally, the methodology can be applied to creating culturally resonant websites, thus supporting New Zealand’s cultural initiatives and promoting cross-cultural understanding. This research lays the groundwork for future studies and has broad applicability across various domains, demonstrating the potential for automated, unbiased website personality classification