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

Browsing School of Engineering, Computer and Mathematical Sciences - Te Kura Mātai Pūhanga, Rorohiko, Pāngarau by Subject "0299 Other Physical Sciences"

Now showing 1 - 6 of 6
  • Item
    A Geometric Approach to Textual Augmented Data Filtering
    (IOP Publishing, 2024-09-09) Feng, SJH; Lai, EMK; Li, W
    Data augmentation is necessary if the amount of training data is insufficient for supervised learning. For natural language processing tasks, obtaining good quality augmented data is not easy. This paper introduces GATFilter, a novel method for filtering out inappropriate augmented textual data for text classification (TC). Utilizing geometric concepts, more specifically the principle component and convex hull analyses, this method adeptly preserves the semantic integrity of words within augmented texts. GATFilter is versatile and applicable across various types of textual augmentation methods. Experiments using several datasets and augmentation strategies showed that classifiers trained with GATFilter-filtered augmented data sets showed improvements in key performance metrics, including accuracy, precision, recall, and F1 score. The method’s efficacy is notably influenced by the quality of the underlying augmentation techniques, indicating its potential to complement and refine various text augmentation strategies. Furthermore, our analysis showed that GATFilter is particularly able to amplify the effectiveness of methods that generate good quality augmented data. GATFilter is openly available online on Github1, and as a Python package2
  • Item
    A Monitoring Campaign (2013–2020) of ESA’s Mars Express to Study Interplanetary Plasma Scintillation
    (Cambridge University Press (CUP), 2023-04-12) Kummamuru, P; Molera Calvés, G; Cimò, G; Pogrebenko, SV; Bocanegra-Bahamón, TM; Duev, DA; Md Said, MD; Edwards, J; Ma, M; Quick, J; Neidhardt, A; De Vicente, P; Haas, R; Kallunki, J; MacCaferri, G; Colucci, G; Yang, WJ; Hao, LF; Weston, S; Kharinov, MA; Mikhailov, AG; Jung, T
    The radio signal transmitted by the Mars Express (MEX) spacecraft was observed regularly between the years 2013-2020 at X-band (8.42 GHz) using the European Very Long Baseline Interferometry (EVN) network and University of Tasmania's telescopes. We present a method to describe the solar wind parameters by quantifying the effects of plasma on our radio signal. In doing so, we identify all the uncompensated effects on the radio signal and see which coronal processes drive them. From a technical standpoint, quantifying the effect of the plasma on the radio signal helps phase referencing for precision spacecraft tracking. The phase fluctuation of the signal was determined for Mars' orbit for solar elongation angles from 0 to 180 deg. The calculated phase residuals allow determination of the phase power spectrum. The total electron content of the solar plasma along the line of sight is calculated by removing effects from mechanical and ionospheric noises. The spectral index was determined as which is in agreement with Kolmogorov's turbulence. The theoretical models are consistent with observations at lower solar elongations however at higher solar elongation ($ ]]>160 deg) we see the observed values to be higher. This can be caused when the uplink and downlink signals are positively correlated as a result of passing through identical plasma sheets.
  • Item
    Anomaly Detection in Text Data Sets Using Character-Level Representation
    (Institute of Physics (IoP), 2021-04-28) Mohaghegh, Mahsa; Abdurakhmanov, Amantay
    This paper proposes a character-level representation of unsupervised text data sets for anomaly detection problems. An empirical examination of the character-level text representation was conducted to demonstrate the ability to separate outlying and normal records using an ensemble of multiple classic numerical anomaly classifiers. Experimental results obtained on two different data sets confirmed the applicability of the developed unsupervised model to detect outlying instances in various real-world scenarios, providing the opportunity to quickly assess a large amount of textual data in terms of information consistency and conformity without knowledge of the data content itself.
  • Item
    Automated Biometric Identification using Dorsal Hand Images and Convolutional Neural Networks
    (Institute of Physics (IoP), 2021-04-01) Mohaghegh, Mahsa; Ash, Payne
    The identification of perpetrators, present in Child Sexual Abuse Imagery (CSAI), is a significant challenge due to the use of anonymisation techniques that mask their identities. Consequently, researchers have investigated the use of uncommon biometric identifiers such as knuckle patterns, palmprints and the dorsal side of the hand. This research proposes a Convolutional Neural Network (CNN) based, fully automated approach to biometric identification using dorsal hand images. The identification performance of three different CNN architectures, AlexNet, ResNet50 and ResNet152, is experimentally determined against two similar datasets, the 11k Hands and IITD dorsal hand databases. A transfer learning approach is used and the final output layers of the CNNs are modified to match the number of classes present in the datasets. The results showed that ResNet CNNs achieved identification accuracies greater than 99.9% on both datasets, whereas the AlexNet CNN achieved between 80.1% and 93.7%. These results demonstrate that it is feasible to use deep, off-the-shelf CNNs, such as ResNets, for automated biometric identification using dorsal hand images. This highlights the potential of using dorsal hand images to identify perpetrators of child sexual abuse from CSAI.
  • Item
    Heat Exchanger Based on Paraffin/Expanded Graphite Composites for Breathing Air Cooling in Fire
    (IOP Publishing, 2021-12-08) Lv, Y; Xiao, J; Huang, Y; Jiang, X; Zhu, Y
    The enormous amount of heat in fires can push inhalation temperature to ~500 K, which is fatal to the civilians. However, conventional rescue respirators are unable to control the breathing air temperature. In this work, we utilized paraffin/expanded graphite (EG) composites to construct a heat exchanger for breathing air cooling. The material itself can be used as the mechanical support, the heat spreader and the heat absorber at the same time. The composites of 0~35 wt% EG were prepared and characterized. The results showed the paraffin was uniformly absorbed in the porous structures of EG. And the paraffin/EG composite with 25 wt% EG has better performance both in simulation and experiment. The heat exchanger constructed by this composite shows good cooling efficiency by cooling the inlet air from 500 K to a breathable 313 K and sustaining for more than 20 minutes.
  • Item
    On the Higher-Order Smallest Ring-Star Network of Chialvo Neurons Under Diffusive Couplings
    (AIP Publishing, 2024-07-18) Nair, Anjana S; Ghosh, Indranil; Fatoyinbo, Hammed O; Muni, Sishu S
    Network dynamical systems with higher-order interactions are a current trending topic, pervasive in many applied fields. However, our focus in this work is neurodynamics. We numerically study the dynamics of the smallest higher-order network of neurons arranged in a ring-star topology. The dynamics of each node in this network is governed by the Chialvo neuron map, and they interact via linear diffusive couplings. This model is perceived to imitate the nonlinear dynamical properties exhibited by a realistic nervous system where the neurons transfer information through multi-body interactions. We deploy the higher-order coupling strength as the primary bifurcation parameter. We start by analyzing our model using standard tools from dynamical systems theory: fixed point analysis, Jacobian matrix, and bifurcation patterns. We observe the coexistence of disparate chaotic attractors. We also observe an interesting route to chaos from a fixed point via period-doubling and the appearance of cyclic quasiperiodic closed invariant curves. Furthermore, we numerically observe the existence of codimension-1 bifurcation points: saddle-node, period-doubling, and Neimark–Sacker. We also qualitatively study the typical phase portraits of the system, and numerically quantify chaos and complexity using the 0–1 test and sample entropy measure, respectively. Finally, we study the synchronization behavior among the neurons using the cross correlation coefficient and the Kuramoto order parameter. We conjecture that unfolding these patterns and behaviors of the network model will help us identify different states of the nervous system, further aiding us in dealing with various neural diseases and nervous disorders.
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.