CoB - SoT&I - Student Works
Permanent URI for this collection
Browse
Recent Submissions
Publication IMMERSIVE DIGITAL AUDIO WORKSTATION(2026-04-16) Hope, Colby; Avelange, Dylan; Ahlers, Eric; Knupp, MichaelThis project is a spatially driven Digital Audio Workstation (DAW) concept that reimagines the typical music creation process by placing the user inside a digital recreation of a studio environment using virtual reality. This project explores how music composition, collaboration, and performance can be made more intuitive and accessible by positioning musical creation tools within a virtual 3D space. Users can interact with instruments and engage with elements of the music creation process through immersive, hands-on experiences in virtual reality. In some cases, this approach can make these tools more accessible to individuals who may not have access to a dedicated mixing or music creation space. Our focus is on emulating as much of the music creation and mixing process as possible while ensuring the environment runs smoothly on standalone VR hardware, usability, and realism.Publication AI CHATBOT FOR PUBLIC FACING BUSINESS WEBSITE(2026-04-16) Shema, Norbert; Desjardins, Michael; Knupp, MichaelThe ScoobySquad AI Agent is an intelligent customer support and workflow automation system designed to enhance the efficiency and scalability of a small service-based business specializing in pet waste removal. This capstone project integrates modern artificial intelligence techniques, including natural language processing and retrieval-augmented generation (RAG), to create a responsive chatbot capable of answering customer inquiries, guiding service requests, and supporting internal operations. The system is built using a combination of web-based technologies, a vector database for semantic search, and automation workflows powered by n8n to connect customer interactions with business logic. At its core, the AI agent leverages a structured knowledge base stored in a cloud database, where frequently asked questions and operational data are embedded into vector representations. When a user submits a question through the website chat interface, the system retrieves the most relevant information using similarity search and generates a context-aware response. This approach allows the AI agent to provide accurate, conversational answers while continuously improving as new data is added. Additionally, n8n enables seamless orchestration of backend processes, such as handling incoming requests, routing data between services, and supporting potential integrations like scheduling and customer management systems. The value of this project lies in its ability to demonstrate how advanced AI technologies can be applied to small, local businesses to improve customer experience and operational efficiency. By reducing the need for constant manual communication, the ScoobySquad AI Agent allows business owners to focus on service delivery while maintaining high-quality customer engagement. This project highlights the growing importance of AI-driven solutions in modern business environments and showcases a practical implementation of intelligent automation.Publication EXTENDED REALITY PROJECTS IN XR277: IMMERSIVE, INTERACTIVE, AND INTEGRATED ENVIRONMENTS(2026-04-16) Nadeau, Alexa; Thy, Anthony; Chaput, Logan; Kennett, Ash; Dorr, Connor; Williams, BraveA collection of student projects from XR277 at Husson University presents extended reality (XR) experiences developed around three core criteria: immersion, interaction, and integration. During the Spring 2026 semester, students created individual projects that are spatially registered in three-dimensional space, enabling users to engage with digital content in a physically coherent environment. Each project supports interaction within three-dimensional space, requiring active user engagement through movement, input, or system response. Projects also incorporate integration by combining real and virtual elements into unified mixed-reality experiences. The work demonstrates a range of applications and design approaches, highlighting how foundational XR principles can be translated into functional experiences. Through iterative prototyping, students addressed challenges related to spatial computing, user interaction, and real-time responsiveness. Collectively, these projects illustrate the potential of XR as a medium for immersive and interactive experiences while emphasizing the role of hands-on development in building technical and conceptual understanding. This showcase reflects how students apply core XR principles to create meaningful, real-time applications within an academic setting.Publication EARLY WARNING FOR CAMPUS CYBER RISKS(2026-04-16) Mwiseneza, Elvis; Lagulos, MarlonCyberattacks are a growing problem for colleges and universities. A single phishing email, stolen password, or infected device can interrupt classes, expose private records, and create major costs for colleges and universities. This project presents XCampus cyber risk monitor, a small lightweight cybersecurity risk monitor designed to give early warning signs before a major or minor incident happens. Instead of only detecting attacks after they begin, the system looks for behavior patterns by learning the pattern and that shows risk is increasing over time. It analyzes six main types of activity such as login failures, suspicious IP access, unusual file downloads, new device connections, phishing email patterns, and privileged access attempts from users. Because real campus security data is private, the project uses realistic designed synthetic data to simulate activity across departments such as finance, admissions, registrar, library, research, and student services. The data is cleaned and grouped by department and system, then transformed into features such as off-hours activity, sudden event spikes, and changes in external access. These features are used in three approaches which are; a machine learning model for risk prediction, an anomaly detection model for unusual behavior, and a rule based baseline for comparison. The system combines these outputs into a final risk score and classifies each department or system as low, medium, or high risk. Results from the prototype, it shows that the predictive model performed better than the rule based method, while anomaly detection helped to show suspicious patterns. The final product includes a simple dashboard for viewing risk levels and trends. This study shows that a practical and low cost early warning system can help universities move from reactive security toward proactive cyber risk management.