Research Bank
Not a member yet
5195 research outputs found
Sort by
SMS phishing detection using machine learning and deep learning techniques
No full textRESEARCH QUESTIONS
How effective are ML and DL models in detecting SMS phishing in New Zealand’s mobile system?
• What impact does class imbalance have, and how can SMOTE address this in phishing detection?
• Which ML and DL models perform best for SMS phishing detection in New Zealand?
• How can ethical data handling be maintained in SMS phishing detection?
• What real-world challenges (model drift, adversarial attacks) could affect SMS phishing detection, and how can they be mitigated?
ABSTRACT
Short Message Service (SMS) is still a vital communication tool in our daily life activities, even with the quick development of Internet protocol-based messaging services. An increasingly sophisticated cyber threat known as SMS phishing (smishing) has emerged in tandem with the rise in mobile device use. As a result, people are finding it hard to distinguish good messages from bad ones. The attackers' propensity for always developing methods has made smishing detection problematic for typical phishing detection techniques including heuristic, feature-based, rule-based, and blacklist approaches. The aim is to construct a New Zealand domain-specific SMS phishing detection system to overcome these difficulties and improve mobile system cybersecurity. The goal is to build an effective ML and DL-based model to accurately detect and categorize SMS smishing messages, addressing class imbalance and ensuring ethical data handling for improved cybersecurity. This study collects the dataset, which is the combination of SMS Smishing Collection from Kaggle and smishing messages from the New Zealand Department of Internal Affairs (DIA) anti-scam archive, ensuring relevance to the local context. The Pre-processing methods involved steps to manage missing and duplicated values, while checking label uniqueness, and performing text pre-processing and lemmatization, followed by label encoding. The dataset is balanced with SMOTE. Random Forest and XGBoost, CNN, RNN, and LSTM are some of the deep learning and machine learning classification models selected for their exceptional performance in text analysis. The models work well for detecting fake SMS messages in the setting of mobile communication networks. Accuracy, precision, recall, and F1score were some of the important measures used to assess the models' performance. The result showed that the XGBoost classifier achieving a superior accuracy of 97.05% compared to other models. This study highlights the practical implications of smishing detection, particularly in real-world mobile communication systems, emphasizing the importance of integrating these models into mobile security applications. Additionally, the research discusses potential future work, including the integration of transformer-based models, the handling of model drift, and addressing adversarial concerns in dynamic environments
Combating decision fatigue and optimising decision-making in Information Technology (IT): The role of Artificial Intelligence (AI) in enhancing cognitive outcomes for IT professionals
No full textDecision fatigue has become a prevalent risk that influences information technology (IT) professionals in New Zealand, affecting the quality of decisions, productivity and overall well-being. This research explored how IT professionals in New Zealand experience decision fatigue and investigated how artificial intelligence (AI) could be strategically utilised to mitigate it. Decision fatigue has been acknowledged in various fields, such as healthcare and psychology, but how it affects IT professionals is under-explored, particularly in IT in New Zealand. As a supportive tool, AI and its usage are explored in different contexts to improve productivity and efficiency, but the specific AI usage to prevent decision fatigue of IT professionals remains unexplored in New Zealand. This research identified this as a gap and an opportunity to explore more in this context. This research aims to explore how IT professionals in New Zealand currently experience decision fatigue and how this can be mitigated using AI. Specifically, it explores; a) the primary challenges in decision-making that lead to decision fatigue among IT professionals, b) the potential benefits and challenges associated with the adoption of AI tools in decision-making processes, and c) how the adoption of AI tools influences their cognitive workload and contributes to the mitigation of decision fatigue.
The research adopted a qualitative interpretivist research design which included conducting semi-structured interviews with fifteen IT professionals who worked in different roles throughout New Zealand. The research used thematic analysis to identify major patterns and insights. In alignment with the research questions, the research first identifies the significant challenges IT professionals face which contribute to decision fatigue. These include cognitive overload, the pressure of multitasking in a time-constrained environment and high-stakes decision-making pressure. Secondly, the research identifies how IT professionals utilise AI in decision-making, such as in automating routine tasks and information processing to support decisions. It further looks at the key limitations of this, such as trust, security and lack of contextual understanding. Finally, strategic approaches were identified, such as automating time-consuming and routine tasks, delegating repetitive decisions, and reducing complexity by processing large amounts of data.
The research provides new perspectives about AI assistance for cognitive well-being through its reduction of decision fatigue in IT settings. The research presents AI-assisted task delegation as a method to maintain mental capacity for complex decision-making tasks. The research emphasizes the need for trust and contextual understanding and human-AI collaboration. The research results guide both decision fatigue theory development and practical AI implementation strategies that support sustainable IT professional workflows with ethical considerations
The impact of the COVID-19 pandemic on the New Zealand-listed companies’ dividend policy
No full textDividends motivate investors to allocate funds to the stock market. The COVID-19 pandemic posed significant challenges to the global economy, impacting companies' financial performance and dividend policies worldwide. This research analyses dividend changes during the COVID-19 pandemic in New Zealand. Employing a quantitative research design, secondary data were sourced from Refinitiv and verified against annual reports. Inferential and descriptive analysis methods, including correlation and panel data regression, examined financial data from 116 publicly listed New Zealand companies between 2017 and 2023. This research findings indicate that New Zealand’s stock market exhibited volatility during the COVID-19 pandemic, with some companies maintaining consistent dividend payouts. This research provides critical insights for investors seeking to preserve the value of their investments by identifying companies that support their value during crises. This research found that New Zealand-listed companies did not distribute dividends uniformly during such periods; some offered higher dividends and enhanced dividends during the COVID-19 pandemic. Interestingly, companies prioritising gender diversity are recognised as value creators. Moreover, listed companies based solely in New Zealand tend to pay higher dividends than those with international operations. This research provides insights into the factors that inform investment decisions, particularly for shareholders seeking strong dividend yields. This research is focused on New Zealand-listed companies which may limit the generalisability of findings on non-listed small and middle-sized companies. This research helps managers and shareholders understand the consequences of the COVID-19 pandemic on dividend policies and highlights the need for capital preservation for the companies' financial stability
Flight of the pīwakawaka: Supporting tertiary learning for Māori with ADHD
No full textWhat is ADHD
Māori & ADHD
Pīwakawaka metaphor
Neuro Affirming Futures
Supporting ākonga aroreretini
He rerekē, he ahurei
He auahatanga
He urutau
He kaitaiki, he toa
He manawa ora
He tirohanga whānui
He hīanga
Reference
Measuring factors impacting Migrant Healthcare Workers (MHCWs) experiences in New Zealand Healthcare
No full textMigrant Healthcare Workers (MHCWs) now constitute a vital segment of Aotearoa New Zealand’s workforce. However, their overall satisfaction with both internal workplace conditions and external social integration following arrival remains under-examined. This study measures the workplace satisfaction of MHCWs and identifies organisational and personal factors that shape their decision to remain in practice. Guided by Herzberg’s Two-Factor Theory, Locke’s Value Theory, and cross-cultural management literature, a survey was administered to MHCWs employed across New Zealand’s public hospitals, aged-care facilities, general practices, and private hospitals. 43 valid responses were received and then analysed.
Findings show a pronounced satisfaction gap between public hospital staff (M= 4.02) and aged-care workers (M=3.16), using a Point 5 Likert-Scale. Overseas qualification recognition yielded a 0.57-point increase in mean satisfaction, underscoring the motivational power of professional validation. Organisational support (r=.94) and social integration (r=.62) emerged as the strongest correlates of overall satisfaction, whereas length of New Zealand tenure showed only a weak association (r=.20). External stressors, especially visa complexity and housing affordability, suppressed satisfaction despite otherwise moderate ratings for work–life balance and community belonging.
The study contributes empirically grounded insight into the drivers of MHCWs retention and offers practical levers for managers: resource audits in aged care, streamlined credential pathways, culturally intelligent leadership, and structured peer integration. Limitations include a small, culturally skewed sample, convenience recruitment, and reliance on basic analysis. Future research should adopt longitudinal and mixed methods designs to track satisfaction across critical career milestones and deepen understanding of under-represented migrant groups
Exploring Te Kooti’s 1886 Te Umutaoroa prophecy as a climate adaptation framework: Prophetic reflections inspired by the Toi Rito Toi Rangatira programme
No full textThis paper is the beginning point of reflecting on and exploring the 1886 Te Umutaoroa prophecy, given by Te Kooti Arikirangi Te Tūruki, as a climate adaptation framework, inspired by the Toi Rito Toi Rangatira - Rangatahi Climate Leadership Programme. Selected by my hapū, Patuheuheu and Ngāti Haka, I joined this programme aimed at rangatahi Māori involved in Deep South Research projects. Despite initial reservations about my age, my kaumātua encouraged my involvement, seeing potential benefits for our hapū. The programme was transformative, enriching my understanding of climate leadership and integrating contemporary perspectives with traditional wisdom. This experience led to a deeper exploration of the Te Umutaoroa prophecy, promising restoration of land, dignity, and sovereignty. The prophecy’s eight mauri provide a foundation for a hapū-centred climate adaptation framework. Climate change, driven by industrialisation, threatens ecosystems, and Māori communities, are particularly vulnerable. Integrating mātauranga Māori with scientific methods offers holistic, culturally meaningful solutions. Te Umutaoroa’s principles—spirituality, land stewardship, hapū well-being, faith/belief, healing, discovering hidden potential, conflict resolution, and returning to ancestral lands—guide potential adaptation strategies. This paper presents Te Umutaoroa as a framework to strengthen the resilience of Patuheuheu and Ngāti Haka against climate change, advocating for further research and collaboration to refine and implement these strategies, ensuring they align with hapū values
Exploring the necessity of corporate dynamic capability and sustainable performance
No full textBACKGROUND
Many studies navigated corporate dynamic capability’s direct and indirect effects on sustainable performance. However, there is an empirical literature conundrum regarding the sufficient and necessary condition of corporate dynamic capability that can enhance the sustainable performance of microfirms.
AIM
The main goal of this study is to explore the necessary and sufficient conditions for corporate dynamic capabilities that support sustainable performance and to determine if agility mediates the relationships between knowledge sharing, managerial cognitive capabilities, sensing capabilities, and sustainable performance.
SETTING
This study surveyed 440 Tanzanian dairy microfirms between July 2021 and January 2022.
METHOD
Partial least squares-path modeling (PLS-PM) and necessary conditions analysis (NCA) were applied from the data collected from 602 managers and employees of microfirms in Tanzania.
RESULTS
The study confirmed that a higher degree of agility, and sensing capability among employees is necessary and sufficient for achieving sustainable performance of microfirms. Agility possibly confounds the relationships between sensing capability, knowledge sharing and managerial cognitive capability on sustainable performance.
CONCLUSION
Knowledge sharing, agility and sensing capability are crucial for sustainable performance, while other dynamic capability factors are not critically sufficient and necessary for growth.
CONTRIBUTION
The study offers valuable insights by identifying critical levels of agility, knowledge sharing, managerial cognitive capability, and sensing capability sufficient for sustainable performance in dairy microfirms. It also provides empirical evidence of agility’s complementary role in mediating the effects of knowledge sharing, sensing capability, and managerial cognitive capability on sustainable performance
Classifying causes of depression from social media posts using machine learning and NLP
No full textRESEARCH QUESTIONS
1 How accurately can machine learning models (e.g., SVM, XGBoost) classify causes of depression expressed in social media posts?
2 Which feature representation—TF-IDF or contextual embeddings (e.g., BERT)—yields better performance for cause classification?
3 How does model performance differ when trained on expert-labeled data versus publicly available self-reported depression datasets?
4 What are the advantages and limitations of using expert-annotated data for classifying causes of depression in terms of accuracy and generalizability?
5 Does combining expert and public datasets improve the robustness and reliability of cause classification models?
ABSTRACT
Depression is a serious challenge to one’s mental health worldwide, affecting billions of souls and causing grievous personal, social, and economic consequences. Detecting de pression early through social media platforms has been a matter of recent interest in the research domain. However, this fails to distinguish between general depression and the underlying cause for it. This is a significant oversight since a therapeutic interven tion, when oriented towards a specific cause like trauma, stress, gender discrimination, or domestic violence, tends to produce far better results. The advancement in machine learning (ML) and natural language processing offers an excellent opportunity to analyze large-scale social media data to detect mental health indicators. This leaves a massive gap, however, in employing these technologies to detect the causes of depression, espe cially with expert-verified standards. This paper considers a new framework for defining depression based on the causes of social media posts. Two complementary datasets are integrated into the study: (1) a small, expert-classified, high-quality dataset annotated by mental health professionals under DSM-5 guidelines, and (2) a large, publicly available dataset of self-disclosed depressive posts. Feature extraction was done with TF-IDF and BERT contextual embeddings. Classification was done with supervised learning through SVM and XGBoost, while latent structures were discovered with unsupervised learning through K-Means. The results indicate that the merged dataset performed better than individual sources, with XGBoost + BERT embeddings achieving the best accuracy and F1 score. Interestingly, unsupervised clustering highlighted latent patterns compatible with known depression causes. The importance of merging expert knowledge with broader social data is thus confirmed by these results, along with the provision of a scalable and interpretable mental health monitoring method.
The study contributes to viewing depression not from a general classification but more specifically, from cause classification, thus allowing for more targeted and timely support and intervention
The identification and spread of the invasive Himalayan wineberry (Rubus ellipticus var. obcordatus) in Albany, Auckland
No full textFirst recorded in Albany, Auckland, New Zealand in 2019, the highly invasive Himalayan wineberry (Rubus ellipticus var. obcordatus (Franch.) Focke), which has caused significant issues in other invasive regions such as Hawaii, is raising concerns about its potential spread and impact in Aotearoa/New Zealand. Due to the similarity of identification of this species compared with other Rubus species, this study aims to provide clear identification characteristics for the species and survey selected areas in the Albany area to determine the potential growth and spread of this invasive weed. In addition, species geospatial modelling was used to predict the potential invasive range within New Zealand for two contrasting trajectories of global warming (RCP 2.6 and 8.5) by the year 2100. The ground survey was completed on foot using 10 meter-wide spaced transects within bush/scrub areas based on predicted occurrences and presence coordinates recorded in Fieldmaps software. Himalayan wineberry presence data showed a higher number of seedlings compared to adult plants and as expected, were predominantly in disturbed areas near other invasive plant species. The results from this modelling found that the main areas at risk of potential invasion were Auckland, Bay of Plenty, Christchurch and Nelson regions. Due to the potential national spread of Himalayan wineberry and the current limited spread in the Albany area, eradication is thought to be feasible, however, monitoring in other regions, particularly the identified hotspots, is an important action for the future
Caloplaca johnwhinrayi S.Y.Kondr. & Kärnefelt (Teloschistaceae) on the Chatham Islands – a new record for Aotearoa / New Zealand
No full textNotice of the presence of Caloplaca johnwhinrayi S.Y.Kondr. & Kärnefelt (Teloschistaceae), previously considered an Australian endemic, is given. The species is known from a single collection from limestone rocks on Motuhinahina Island within Te Whanga Lagoon, Rēkohu / Wharekauri / Chatham Island (the largest of the Chatham Islands group), Aotearoa / New Zealand. Brief notes on the species recognition, ecology and conservation are given, with the recommendation that it should be looked for in similar habitats elsewhere in Aotearoa / New Zealand