Welcome to Research Commons

Research Commons is the University of Waikato's open access research repository, housing research publications and theses produced by the University's staff and students.

Communities in Research Commons

Select a community to browse its collections.

Now showing 1 - 2 of 2

Recent Submissions

A physics-informed kernel approach to learning the operator for parametric PDEs
(Springer Nature, 2024) Kurz, Jason A.; Bowman, Brett A.; Seman, Matthew G.; Oian, Chad A.; Khan, Taufiquar
Operator networks are neural networks designed to learn operators with special emphasis on solution operators for parameterized families of partial differential equations (PDEs). Once trained, operator networks can provide a solution to a PDE more quickly than current numerical PDE solvers by several orders of magnitude. Fourier neural operators (FNOs) and deep operator networks (DeepONets) are the two primary operator networks in existence for learning the solution operator to PDEs and have mostly only been applied to two-dimensional or three-dimensional problems, due to the computational expense of training networks in higher dimensional settings. The sole exception is a model-parallel FNO, which decomposes the function input domain space. We demonstrate a neural operator network with a physics-informed integral kernel that, once trained, is able to predict skin and ocular media’s time-dependent thermal response to incident laser radiation much more rapidly than existing numerical algorithms.
NFA: A neural factorization autoencoder based online telephony fraud detection
(Elsevier, 2024) Wahid, Abdul; Msahli, Mounira; Bifet, Albert; Memmi, Gerard
The proliferation of internet communication channels has increased telecom fraud, causing billions of euros in losses for customers and the industry each year. Fraudsters constantly find new ways to engage in illegal activity on the network. To reduce these losses, a new fraud detection approach is required. Telecom fraud detection involves identifying a small number of fraudulent calls from a vast amount of call traffic. Developing an effective strategy to combat fraud has become challenging. Although much effort has been made to detect fraud, most existing methods are designed for batch processing, not real-time detection. To solve this problem, we propose an online fraud detection model using a Neural Factorization Autoencoder (NFA), which analyzes customer calling patterns to detect fraudulent calls. The model employs Neural Factorization Machines (NFM) and an Autoencoder (AE) to model calling patterns and a memory module to adapt to changing customer behaviour. We evaluate our approach on a large dataset of real-world call detail records and compare it with several state-of-the-art methods. Our results show that our approach outperforms the baselines, with an AUC of 91.06%, a TPR of 91.89%, an FPR of 14.76%, and an F1-score of 95.45%. These results demonstrate the effectiveness of our approach in detecting fraud in real-time and suggest that it can be a valuable tool for preventing fraud in telecommunications networks.
COVID-19 muddles talanoa and vā: Perceived connections and uncertainties
(Wilf Malcolm Institute of Educational Research, New Zealand, 2021) Laulaupea’alu, Siuta
Learning online from home bubbles through the use of information communication technology (ICT) stretches the engagement and enactment of vā (relational connections) between students and lecturers as well as Pacific people in the community. In this paper, talanoa is used to capture students’ online learning experiences and their perceived understanding of connections. Such experiences are embodied in people’s interactions, conversations, problem-solving, knowledge sharing and exchange of ideas and practice. As the vā space online between lecturer and student as well as people in the community is physically mama’o (distanced), the perceived space of learning connection raises concern over ethics and practice. Engaging in open talanoa of the uncertainties linked to online interactions within the post-COVID context and the place of vā ethics can lead to talanoa mālie that highlight possibilities and solutions.
Linear adaptive filtering for regression in data streams
(SpringerNature, 2025-05-06) Igual, J; Gomes, HM; Pfahringer, B; Bifet, A
Many applications in supervised learning of evolving data streams deal with regression, since they need to forecast numeric values in a fast and accurate way. Most of the current regression methods are inspired by the state-of-the-art data stream classifiers adapted to the regression problem. In this paper, we present a novel approach to the problem based on classical linear adaptive filtering theory. The linearity allows to obtain simple general models highly accurate without needing to use ensemble learners. We study the recursive version of these methods as they are the ones that satisfy the stream requirements and we include in their formulation the ability to track the drift of the data in stream scenarios. We show how they integrate in an easy and fast way the adaptation to changing environments avoiding the use of explicit drift detectors. We apply these methods to classical datasets in stream regression and show how they can outperform them in drifting cases when the linear model is a good approximation to the problem.
Proximal near-infrared hyperspectral imaging dataset for identifying epicuticular wax loss in Masena blueberries to evaluate post-harvest quality
(Elsevier, 2025-08-05) Faisal, S; Thawdar, Y; Ooi, MPL; Reutemann, P; Fletcher, D; Kuang, YC; Abeysekera, SK
The dataset presented in this paper consists of hyperspectral images of Masena blueberries that were harvested on November 24, 2023, from an orchard in Pukehina, New Zealand. Blueberries were hand-harvested with gloves (intact wax), hand-harvested (without gloves), and mechanically aided by picking via a handheld shaker. Some berries were also wiped to eliminate degrading epicuticular wax (EW) for comparison. Imaging was performed within 9 hours of harvest using a Specim FX17e hyperspectral camera (900–1700 nm, 224 bands) under controlled lighting conditions. The data were white and dark reference-normalized, annotated using the in-house HAPPy tool (ENVI Software), and saved in MATLAB (.mat) format for analysis. A total of 49 individual hyperspectral images were captured from 39 blueberry fruits to capture multiple views or surface states. We provide 5 spectral hypercube sets of data collected with the hyperspectral camera: ‘Assisted Harvested Blueberries (AHB)’ (10 images), ‘Hand Harvested Blueberries (HHB)’ (10 images), ‘Perfect EW’ (10 images), ‘No EW’ (9 images), and ‘No EW vs. Perfect EW‘ (10 images: 5 from ‘No EW‘ and 5 from ‘Perfect EW‘). This dataset, collected and archived by the University of Waikato (WaI2M: Waikato Instrumentation and Measurement Research Group, Hyperspectral Imaging Group), enables near-infrared hyperspectral imaging research in agriculture for EW classification and detection, harvesting method classification, and fruit surface property spectral analysis using machine/deep learning methods.