Lakehead University Knowledge Commons

Knowledge Commons is an open access repository for scholarship and research produced at Lakehead University. It is a free and secure repository for LU faculty, students, staff, and researchers to preserve and present their scholarship.

Communities in KnowledgeCommons

Select a community to browse its collections.

Now showing 1 - 3 of 3

Recent Submissions

An integrated framework for art image novelty detection and ownership tracing using deep Siamese networks and blockchain
(2026) Zhou, Yutao; Rathore, M. Mazhar; MacNeil, Moira; Deng, Yong
Digital artworks are increasingly distributed and shared across online platforms, raising critical challenges in both ownership verification and similarity detection. On one hand, it is difficult to establish secure, tamper-resistant records of artwork ownership in decentralized environments. On the other hand, identifying whether a newly submitted artwork is visually similar to an existing one remains a non-trivial task, especially under various artistic transformations such as style transfer, inpainting, and compositional editing. Vision Models can effectively compare image content, but they lack mechanisms to securely protect ownership. In contrast, blockchain technologies offer immutability, traceability, and decentralized data storage, yet lack the capability to evaluate visual similarity. These limitations highlight the need for an integrated solution that jointly addresses both visual similarity detection and secure ownership verification. To resolve those issues, this thesis proposes a blockchain-based artwork verification system that integrates deep visual similarity detection with decentralized ownership registration. In the proposed framework, blockchain is used to register artwork ownership and store compact image feature payloads as immutable on-chain records, while offchain deep learning models extract visual embeddings and perform similarity matching. Multiple visual models are trained and evaluated under different distance metrics, loss functions, and Siamese architectures. To improve the practicality of on-chain storage, the extracted embeddings are projected into lower dimensions, quantized into compact payloads, and then analyzed for storage feasibility and matching performance. The blockchain component is further evaluated through experiments on payload storage, update cost, retrieval efficiency, and large-scale matching simulation. Experimental results show that the ResNet50 model trained with NT-Xent loss and Euclidean distance achieves the best overall performance among the tested settings, while DeiT-small performs competitively at higher embedding dimensions. The results further indicate that quantized and compressed embeddings can significantly reduce blockchain storage cost while preserving most retrieval capability, although lower-dimensional embeddings increase the false-positive rate in the final similarity simulation.
Domain adaptation for robust WiFi sensing: empirical analysis of domain shift and class imbalance in WiFi CSI-based human activity recognition
(2026) Nowdeh, Alireza Rajoli; Shirehjini, Ali Asghar Nazari
WiFi-based human activity recognition (HAR) has emerged as a promising device-free sensing technology for smart homes, healthcare monitoring, and ambient assisted living. However, these models suffer from performance degradation when applied to new environments because of domain shift caused by changes in room layout, multipath propagation, and line-of-sight (LOS) versus non-line-of-sight (NLOS) conditions. This thesis addresses the robustness gap by extending Dual Adversarial Network for Human Activity Recognition (DA-HAR) with two components: a Conditional Domain Adversarial Network (CDAN) for aligning joint feature and prediction distributions and a class-weighted learning strategy to mitigate the effects of class imbalance. The proposed framework is evaluated on a public WiFi CSI dataset with 12 fine-grained activities collected in three environments (corridor, office, classroom) under LOS and NLOS conditions. Comprehensive experiments are conducted using three transfer scenarios (E1,E2→E3 / E1,E3→E2 / E2,E3→E1). The results indicate that the enhanced DA-HAR consistently outperforms both the original DA-HAR and the source-only baseline, achieving up to 7.1% absolute and 17.2% relative accuracy improvements. The confusion matrix shows that DA-HAR + CDAN reduces misclassification among fall-related and locomotion activities, indicating that it preserves class structure under domain shift better. To isolate the effect of class imbalance, an experiment is performed on 20 random six-activity subsets with and without the class-weighted loss. Class weighting improves accuracy from 0.6806 to 0.7112 and increases CDAN improvement from 5.3% to 8.7%, which means that class imbalance negatively impacts both the source model and the adaptation process. However, performance for fall activities depends on the scenario, and bending is still difficult to recognize, indicating remaining challenges. This work provides empirical evidence that robust domain adaptation for WiFi sensing requires both conditional alignment of class-specific distributions and explicit treatment of class imbalance. The proposed method offers a principled step toward reliable activity recognition in real-world environments.
Multicultural educational resources in Thunder Bay Public Libraries: availability, accessibility, and educational implications
(2026) Jin, Mengyue; Kaefer, Tanya; Gonzalez, Ismael
This study examines the availability, accessibility, and perceived usefulness of multicultural educational resources at one branch of the Thunder Bay Public Libraries, with particular attention to how these resources support multicultural families with young children. As Canadian mid-sized cities become increasingly diverse, public libraries play an important role in supporting bilingual literacy development, heritage language maintenance, and opportunities for cultural learning. However, limited research has explored how multicultural educational resources function in mid-sized cities with constrained institutional capacity. Using a qualitative single-case study design informed by ethnographic principles, this study draws on semi-structured interviews with three parents from diverse linguistic backgrounds and two library staff members, as well as field observations of library spaces and resources displays. Data were analyzed through thematic analysis to identify patterns related to resource availability, visibility, institutional practices, and family experiences. Findings indicate that while the library provides a welcoming and child-centered environment that supports English literacy and social interaction, multicultural educational resources remain limited in scope, visibility, and depth. Multilingual materials are unevenly distributed and often difficult to locate without staff assistance. Although families value the library as a community space, support for sustained heritage language development and culturally embedded learning is constrained by institutional structures, centralized decision-making, and limited programming opportunities. The study highlights the importance of visibility, strategic collection development, community-informed programming, and institutional support in advancing equitable access to multicultural educational resources. By focusing on a mid-sized Canadian city, this research contributes to ongoing discussions about the role of public libraries as inclusive educational spaces in increasingly diverse communities.
Evidence-grounded clinical pharmacogenomics question answering system using large language models and hybrid retrieval augmentation
(2026) Arafin, Protiva; Alkhateeb, Abedalrhman; Moniruzzaman, Md; Alsmadi, Malek; Ahmed, Saad B.
Pharmacogenomics (PGx) is very important for personalized medicine since it helps doctors choose the right drugs and doses based on a person’s genetic makeup. But the growing amount and complexity of PGx data, as well as the requirement to understand clinical recommendations, make it harder to make good decisions. This study puts forward a data-driven clinical decision support framework that combines large language models (LLMs) with hybrid retrieval-augmented generation (RAG) to enhance the response to pharmacogenomic questions. The framework assesses two contemporary LLMs, Meta-LLaMA-3.1-8B-Instruct and Qwen3-8B, through various configurations, encompassing base models, Low-Rank Adaptation (LoRA) fine-tuning, and hybrid RAG-based methodologies. The structured pharmacogenomics data from CPIC and the clinical guideline information from ClinPGx are combined to make a huge dataset. To make it easier to find and use in models, the data goes through procedures including merging, cleaning, normalizing, and converting to JSONL format.A hybrid retrieval approach is aimed to enhance factual grounding by integrating lexical filtering with semantic similarity through sentence embeddings. This research use both automatic metrics and manual checks to rate the models on their correctness, relevance, completeness, and clarity. The results reveal that Qwen works well as a basic model, and that LLaMA gets much better when it is used with RAG and LoRA, giving answers that are more aware of the context and therapeutically useful. Fine-tuning alone doesn’t always work, which shows how limited it is to only use parametric data. The results show that accuracy in clinical settings needs to be backed up by consistency, relevance, and evidence. This study demonstrates that employing retrieval methods alongside parameter-efficient fine-tuning enhances the reliability and utility of LLM-based systems in clinical environments. The proposed methodology establishes a scalable framework for the development of trustworthy AI-driven solutions in pharmacogenomics and healthcare decision support.
Environmental education in healthy active living education courses: a justice-oriented resource for Ontario secondary educators
(2026) Phillips, Madison M.; Russell, Connie; Kerr, Donald
The purpose of this portfolio is to develop a resource that supports Ontario Healthy Active Living Education (HALE) teachers in integrating environmental education (EE) through a justice-oriented lens. While curriculum documents may highlight the importance of EE, they often offer limited guidance on how to meaningfully apply it within Health and Physical Education (HPE) contexts. This resource aims to bridge that gap by connecting environmental issues to health, wellbeing, and social justice. There are three tasks in this portfolio. The first task is a literature review that examines how EE is defined and connected to human health, while situating it within broader social justice frameworks. It analyzes Ontario curriculum and policy documents to assess how EE is positioned within the HPE curriculum, establishing the foundation for the subsequent components of this portfolio. The second task is a digital resource in the form of a website that is designed to support HALE educators by curating relevant articles and resources. The site also features annotated summaries that outline a foundational article’s key argument and its relevance to EE, along with considerations for classroom application, to bridge the gap between theory and practice. The final task is a short reflection on the portfolio process and the portfolio's effort to challenge dominant ideologies and engage with diverse ways of knowing within HPE and EE.