School of Engineering, Computer and Mathematical Sciences - Te Kura Mātai Pūhanga, Rorohiko, Pāngarau

Permanent link for this collection

https://hdl.handle.net/10292/553
AUT is home to a number of renowned research institutes in engineering, and computer and mathematical sciences. The School of Engineering, Computer and Mathematical Sciences strong industry partnerships and the unique combination of engineering, computer and mathematical sciences within one school stimulates interdisciplinary research beyond traditional boundaries. Current research interests include:
  • Artificial Intelligence; Astronomy and Space Research;
  • Biomedical Technologies;
  • Computer Engineering; Computer Vision; Construction Management;
  • Data Science;
  • Health Informatics and eHealth;
  • Industrial Optimisation, Modelling & Control;
  • Information Security;
  • Mathematical Sciences Research; Materials & Manufacturing Technologies;
  • Networking, Instrumentation and Telecommunications;
  • Parallel and Distributed Systems; Power and Energy Engineering;
  • Software Engineering; Signal Processing; STEM Education;
  • Wireless Engineering;

Browse

Recent Submissions

Now showing 1 - 5 of 1408
  • Thumbnail Image
    Item
    A Survey on Graph Neural Network-Based Next POI Recommendation for Smart Cities
    (Springer Science and Business Media LLC, 2024-07-26) Yu, J; Guo, L; Zhang, J; Wang, G
    Amid the rise of mobile technologies and Location-Based Social Networks (LBSNs), there’s an escalating demand for personalized Point-of-Interest (POI) recommendations. Especially pivotal in smart cities, these systems aim to enhance user experiences by offering location recommendations tailored to past check-ins and visited POIs. Distinguishing itself from traditional POI recommendations, the next POI approach emphasizes predicting the immediate subsequent location, factoring in both geographical attributes and temporal patterns. This approach, while promising, faces with challenges like capturing evolving user preferences and navigating data biases. The introduction of Graph Neural Networks (GNNs) brings forth a transformative solution, particularly in their ability to capture high-order dependencies between POIs, understanding deeper relationships and patterns beyond immediate connections. This survey presents a comprehensive exploration of GNN-based next POI recommendation approaches, delving into their unique characteristics, inherent challenges, and potential avenues for future research.
  • Thumbnail Image
    Item
    A Risk Assessment Framework for Mobile Apps in Mobile Cloud Computing Environments
    (MDPI AG, 2024-07-29) Ogwara, Noah Oghenefego; Petrova, Krassie; Yang, Mee Loong; MacDonell, Stephen G
    Mobile devices (MDs) are used by mobile cloud computing (MCC) customers and by other users because of their portability, robust connectivity, and ability to house and operate third-party applications (apps). However, the apps installed on an MD may pose data security risks to the MD owner and to other MCC users, especially when the requested permissions include access to sensitive data (e.g., user’s location and contacts). Calculating the risk score of an app or quantifying its potential harmfulness based on user input or on data gathered while the app is actually running may not provide reliable and sufficiently accurate results to avoid harmful consequences. This study develops and evaluates a risk assessment framework for Android-based MDs that does not depend on user input or on actual app behavior. Rather, an app risk evaluator assigns a risk category to each resident app based on the app’s classification (benign or malicious) and the app’s risk score. The app classifier (a trained machine learning model) evaluates the permissions and intents requested by the app. The app risk score is calculated by applying a probabilistic function based on the app’s use of a set of selected dangerous permissions. The results from testing of the framework on an MD with real-life resident apps indicated that the proposed security solution was effective and feasible.
  • Thumbnail Image
    Item
    Overlapping Shoeprint Detection by Edge Detection and Deep Learning
    (MDPI AG, 2024-07-31) Li, Chengran; Narayanan, Ajit; Ghobakhlou, Akbar
    In the field of 2-D image processing and computer vision, accurately detecting and segmenting objects in scenarios where they overlap or are obscured remains a challenge. This difficulty is worse in the analysis of shoeprints used in forensic investigations because they are embedded in noisy environments such as the ground and can be indistinct. Traditional convolutional neural networks (CNNs), despite their success in various image analysis tasks, struggle with accurately delineating overlapping objects due to the complexity of segmenting intertwined textures and boundaries against a background of noise. This study introduces and employs the YOLO (You Only Look Once) model enhanced by edge detection and image segmentation techniques to improve the detection of overlapping shoeprints. By focusing on the critical boundary information between shoeprint textures and the ground, our method demonstrates improvements in sensitivity and precision, achieving confidence levels above 85% for minimally overlapped images and maintaining above 70% for extensively overlapped instances. Heatmaps of convolution layers were generated to show how the network converges towards successful detection using these enhancements. This research may provide a potential methodology for addressing the broader challenge of detecting multiple overlapping objects against noisy backgrounds.
  • Thumbnail Image
    Item
    Multi-objective Optimisation Framework for Standalone DC-Microgrids with Direct Load Control in Demand-Side Management
    (Institution of Engineering and Technology (IET), 2024-07-30) Jayasinghe, H; Gunawardane, K; Zamora, R
    Renewable energy-powered DC microgrids have emerged as a sustainable alternative for standalone power systems in remote locations, which were traditionally reliant on diesel generators (DIG) only. To ensure power quality and reliability, energy storage systems (ESS) and demand-side management (DSM) techniques are employed, addressing the intermittent nature of renewable energy sources (RES). This manuscript presents a novel multi-objective optimisation framework to determine the equipment sizing, depth of discharge (DoD) of ESS, and share of controllable loads contributing to DSM in a standalone DC microgrid incorporated with RES as a primary energy source and a backup DIG. The proposed optimisation strategy utilises genetic algorithm with the objectives of minimizing lifecycle cost and carbon footprint. A novel battery energy storage system (BESS) management criterion is introduced, which accounts for battery degradation in the lifecycle cost calculation. The minimum allowable DoD of the BESS is considered a decision variable in the optimisation problem to assess the impact of higher DoD on lifecycle cost improvement. MATLAB simulation results demonstrate that the proposed optimisation model significantly reduces the levelized cost of electricity and per unit carbon footprint compared to previous models. Additionally, it identifies an optimal range of DoD for the BESS to enhance the lifecycle cost of a standalone DC microgrid.
  • Thumbnail Image
    Item
    Swin-Fake: A Consistency Learning Transformer-Based Deepfake Video Detector
    (MDPI AG, 2024-08-01) Gong, Liang Yu; Li, Xue Jun; Chong, Peter Han Joo
    Deepfake has become an emerging technology affecting cyber-security with its illegal applications in recent years. Most deepfake detectors utilize CNN-based models such as the Xception Network to distinguish real or fake media; however, their performance on cross-datasets is not ideal because they suffer from over-fitting in the current stage. Therefore, this paper proposed a spatial consistency learning method to relieve this issue in three aspects. Firstly, we increased the selections of data augmentation methods to 5, which is more than our previous study’s data augmentation methods. Specifically, we captured several equal video frames of one video and randomly selected five different data augmentations to obtain different data views to enrich the input variety. Secondly, we chose Swin Transformer as the feature extractor instead of a CNN-based backbone, which means that our approach did not utilize it for downstream tasks, and could encode these data using an end-to-end Swin Transformer, aiming to learn the correlation between different image patches. Finally, this was combined with consistency learning in our study, and consistency learning was able to determine more data relationships than supervised classification. We explored the consistency of video frames’ features by calculating their cosine distance and applied traditional cross-entropy loss to regulate this classification loss. Extensive in-dataset and cross-dataset experiments demonstrated that Swin-Fake could produce relatively good results on some open-source deepfake datasets, including FaceForensics++, DFDC, Celeb-DF and FaceShifter. By comparing our model with several benchmark models, our approach shows relatively strong robustness in detecting deepfake media.
Items in these collections are protected by the Copyright Act 1994 (New Zealand). These works may be consulted by you, provided you comply with the provisions of the Act and the following conditions of use:
  • Any use you make of these works must be for research or private study purposes only, and you may not make them available to any other person.
  • Authors control the copyright of their works. You will recognise the author’s right to be identified as the author of the work, and due acknowledgement will be made to the author where appropriate.
  • You will obtain the author’s permission before publishing any material from the work.