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

De-Sisters: Understanding women’s desistance from crime in Aotearoa
(The University of Waikato, 2026) Stevens, Shannon; Curtis, Cate; Curtis, Bruce
Desistance from crime is increasingly understood as a developmental and relational process embedded within social environments. Efforts to establish a life beyond crime are frequently complicated by cumulative disadvantages, including unresolved trauma, fractured or harmful relationships, economic precarity, and enduring experiences of systemic exclusion. Despite growing scholarly interest in desistance, women’s lived experiences of navigating change remain marginal within the literature, with limited empirical attention given to the specific social, cultural, and structural contexts shaping desistance in Aotearoa New Zealand. Across the narratives of twelve women once caught in cycles of recidivism, this study offers new insight into the factors which support and obstruct lives beyond crime. Desistance is revealed as a process of self-discovery, healing and relational connection. The study maps the psychological and relational terrain of change, exposing an uneven landscape produced by minimal access to therapeutic intervention and limited cultural integration. Under these conditions, efforts to reconstruct identity were further complicated by the persistent digital trace of conviction, amplifying both private and public forms of stigma. Women leaving prison are often unprepared and underdeveloped, desisting within hazardous reintegration environments. Desistance advanced when agency was restored, highlighting the need for coordinated systemic reform to strengthen human, social, and recovery capital, modernise correctional provisions, and formalise post-release advocacy.
How positioning wearable haptic interfaces on limbs influences virtual embodiment
(IEEE, 2024) Dwivedi, Anany; Yu, Shihan; Hao, Chenxu; Salvietti, Gionata; Prattichizzo, Domenico; Beckerle, Philipp
With increasing use of computer applications and robotic devices in our everyday life, and with the advent of metaverse, there is an urgent need of developing new types of interfaces that facilitate a more intuitive interaction in physical and virtual space. In this work, we investigate the influence of the location of haptic feedback devices on embodiment of virtual hands and user load during an interactive pick-and-place task. To do this, we conducted a user study with a 3x2 repeated measure experiment design: feedback position is varied between the distal phalanx of the index finger and the thumb, the proximal phalanx of the index finger and the thumb, and the wrist. These conditions of feedback are tested with the stimuli applied synchronously to the participant in one case, and with an additional delay of 350 ms in the second case. The results show that the location of the haptic feedback device does not affect embodiment, whereas the delay, i.e., whether the feedback is applied synchronously or asynchronously, affects embodiment. This suggests that for pick-and-place tasks, haptic feedback devices can be placed on the user's wrist without compromising performance making the hands to remain free, allowing unobstructed hand visibility for precise motion tracking, thereby improving accuracy.
Leveraging large language models for automated feature extraction and model training in EMG-based motion decoding
(IEEE, 2025) Dwivedi, Anany; Guan, Bonnie; Lahr, Gustavo J. G.; Head, Mitchell A.; König, Jemma; Owen, Mahonri; Liarokapis, Minas; Bifet, Albert
Feature extraction and model training are critical steps in developing machine learning models for electromyography (EMG) based motion decoding. Traditionally, these processes require domain expertise and programming knowledge to implement signal processing algorithms with optimized model training pipelines. In this work, we investigate the feasibility of using Large Language Models (LLMs) to automate both the extraction of features from EMG data and the development of machine learning models for decoding human motion with minimal human intervention. More specifically, we compare LLM extracted features and their corresponding motion decoding models against those developed using manually developed code. Our results indicate that LLM extracted features and their corresponding trained models achieve performance comparable to traditional methods, demonstrating the potential of accelerating research and scientific investigations with AI-driven biosignal processing. This study highlights LLMs’ capabilities and limitations in replacing manual coding for developing muscle-machine interfaces and provides insights into their integration into biomedical signal analysis workflows.
On the impact of different light wavelengths in decoding human intention in lightmyography controlled prosthetic hands
(IEEE, 2025) Guan, Bonnie; Godoy, Ricardo V.; Dwivedi, Anany; Liarokapis, Minas
Lightmyography is a method of human interfacing proposed for the control of prosthetic systems in which light is reflected on the skin and captured using photosensors. While sharing conceptual similarities with forcemyography, lightmyography differentiates itself in the way it interacts with the skin due to skin optics. Different wavelengths of light interact with the skin in different depths and degrees, a principle utilized in photoplethysmography to estimate heart rate and blood oxygen levels. Therefore, lightmyography was previously designed with two wavelengths of light, green and near-infrared. This study seeks to examine the impact of using different wavelengths of light in the decoding of hand postures using different machine learning methods. Four light configurations were tested: using only green light, using only infrared light, using both green and infrared light blinking alternately each 250 ms, and finally using both lights blinking at 125 ms. Data was decoded using four machine learning algorithms, namely linear discriminant analysis, random forest, convolutional neural networks, and a transformer-based model. Statistical testing showed no significant effect from light configuration in the machine learning decoding of hand posture performance. The dataset collected in this study has been made available for further investigation.
The application of search heuristics to the single machine early/tardy scheduling problem
(The University of Waikato, 1995) James, Ross Jonathan William; Buchanan, John; Foulds, Les R.
Just-In-Time manufacturing aims to reduce waste within an organisation, by focusing on completing orders at the time requested by customers. Failure to complete orders at the time specified causes the organisation to incur additional costs. The Just-In-Time environment has introduced complexities for scheduling which have been modelled by the Single Machine, Early/Tardy Machine Scheduling problem. Research to date has concentrated on models with assumptions such as the use of a single due date for all jobs or restrictions on early/tardy penalties. This research uses a generalised model which makes no such assumptions. The use of three popular ‘intelligent’ search techniques: Simulated Annealing, Genetic Algorithms and Tabu Search are investigated as possible solution techniques. The properties of the common due date, early/tardy scheduling problem enable schedules to be defined from early/tardy job specifications - which state whether individual jobs are scheduled earlier or later than requested. Search techniques can then be applied using an early/tardy job specification solution space rather than the sequence of jobs solution space. Experiments found that the former was more efficient. The properties of the common due date problem are generalised for the distinct due date problem so that a distinct due date schedule can also be defined from early/tardy job specifications, however, this is only possible with the use of a heuristic. By incorporating an early/tardy heuristic into a search, the early/tardy job specification solution space can be used in a search technique. Two early/tardy heuristics are developed, one being a construction based heuristics, the other based on a principle of conflict resolution. Both of these heuristics, unlike any other heuristic for this problem, determine the sequence and timing of each job simultaneously. Experiments demonstrated that the early/tardy heuristics could find good quality solutions, however, search techniques based on the standard sequence of jobs solution space found better solutions. On larger sized problems, searches based on the early/tardy heuristic made large initial gains but little subsequent progress. Hybrid techniques which used both the heuristic solution space and the sequence of jobs solution space were found to produce the best results. Of the search techniques, Tabu Search was found to consistently produce better results than Simulated Annealing and Genetic Algorithms. Experiments also found that a Tabu Search with an early/tardy heuristic-based diversification strategy was the least sensitive to the starting point used by the search. A final experiment found that if an error in the earliness and tardiness penalties is within the range of -25% and 20%, there is unlikely to be a significant difference in the solution quality.