crackmitacsAll disciplines

Accounting

21 Mitacs Globalink (GRI) research projects for Summer 2027.

1. AI-Assisted Financial Trade Reconciliation Using Large Language Models

This project explores how large language models (LLMs) can be applied to financial trading systems to support real-time trade inference and reconciliation. In modern trading environments, discrepancies between executed trades and recorded transactions can persist for several days, creating operational risk and potential exposure to fraud or misreporting. Traditional reconciliation processes rely on structured transaction data and delayed matching, which limits the ability to detect issues early. The aim of this project is to investigate whether unstructured communication, such as trader chat, can be used as an additional signal for identifying discrepancies. The project will involve developing a simulation-based environment that models trading activity, booking errors, and associated communication, allowing controlled experimentation under realistic conditions . Within this environment, the student will explore how LLM-based systems can interpret conversational data to infer executed trades and compare them against recorded transactions. This includes examining how factors such as ambiguity, noise, and communication style affect model performance. The project will also consider how these methods could be integrated into existing accounting and control processes as an additional layer of monitoring. The expected outcome is a prototype system and a set of experimental results that evaluate the effectiveness and limitations of LLM-based approaches in this context. The project contributes to ongoing work at the intersection of financial systems, cybersecurity, and AI, and is relevant to both academic research and industry applications in trading, risk management, and financial control.

Research area, student roles & skills

Research area: My research sits at the intersection of cybersecurity, critical infrastructure, and financial systems, focusing on how complex environments can be made more resilient to disruption and uncertainty. My work has examined high-impact sectors such as energy and operational technology, alongside six years of experience in the financial services industry, which informs my interest in how technical risks translate into business and market outcomes. I am particularly interested in how data-driven systems support decision-making under uncertainty, and how risks can be identified, evaluated, and communicated effectively.

Student roles:
The student will be involved in the early scoping and design of the project, helping to identify relevant use cases for applying large language models within financial trading and reconciliation contexts. This will include thinking through how trading activity, booking errors, and communication can be represented within a simulation environment.

They will contribute to the development of prototype tools, primarily in Python, working on components such as data handling, model interaction, and simulation logic. This may include working with APIs, processing structured and unstructured data, and implementing simple evaluation workflows. The student will work closely with Dr. Higgins throughout the project, with regular discussions to refine ideas, review progress, and guide development.

In addition, the student will assist in testing and evaluating the system, exploring how well LLM-based approaches perform under different conditions, such as noisy or ambiguous inputs. This may involve running experiments, comparing results, and identifying limitations or areas for improvement.

Overall, the role is hands-on and exploratory, with a balance between design, development, and analysis, and an emphasis on building and testing practical systems in a real-world context.

Skills required:
The ideal student will have some experience with programming or software development, with familiarity in Python being particularly useful. A background in computer science, engineering, or a related field would be beneficial, although students from other disciplines with strong technical skills are also encouraged to apply. An interest in data, AI, or financial systems would be helpful.

Some familiarity with accounting, finance, or trading concepts would be an advantage, particularly for understanding reconciliation and financial data, but this is not essential. The project is well suited to students who are comfortable working across technical and applied domains, and who are

2. Best Practices for Developing Next Gen Family Business Leaders

This project investigates what helps young people become well prepared, committed next generation contributors to family enterprises. While many families want the “next gen” to be capable owners or leaders, there is limited evidence on which educational experiences, governance exposure, mentoring pathways, and identity factors truly support that development. This project addresses that gap using the Next Gen Development Study, an ethics approved survey of over 900 undergraduate business students from the Ted Rogers School of Management—the largest undergraduate business cohort in Canada. The dataset includes detailed domains: demographics and family business background; nature of the family firm (where applicable); governance structures and practices; intentions to work in or own the family business; family business education and experiential learning; and measures of commitment, identity, and perceived fit. It also contains open ended reflections on perceived challenges, expectations, and opportunities. The Globalink intern will analyze how combinations of education (courses, experiential programs, extracurriculars), governance exposure (boards, family councils, meetings), and mentoring relationships (inside/outside the family) are associated with stronger readiness and intention to contribute—either in operating roles, ownership roles, or as entrepreneurial collaborators. Quantitative analysis will be complemented by systematic coding of open ended responses to bring student voices into the recommendations. Based at the Family Business Institute in downtown Toronto and supervised by award winning scholar Dr. Francesco Barbera, the intern will help turn these findings into clear best practice guidance for families, educators, and advisors. Outputs may include a practitioner oriented white paper, short briefs, and visual summaries aimed at non academic audiences. The student’s contributions can be publicly recognized via a profile on the FBI website and acknowledgment on reports or presentations, creating a highly visible, CV enhancing applied research experience.

Research area, student roles & skills

Research area: Our research focuses on next generation development in family enterprises, at the intersection of family business, education, organizational behavior, and identity studies. We examine how educational experiences, governance exposure, mentoring relationships, and identity/commitment shape whether and how young people engage with their family’s business—as future owners, managers, or entrepreneurial partners. Using large scale survey data from over 900 business students in Canada’s largest undergraduate business school, we compare students from family business and non family backgrounds across readiness, intentions, and perceived opportunities. The work is grounded in downtown Toronto’s diverse context and emphasizes translating rigorous evidence into practical guidance.

Student roles:
The Globalink intern will play a central role in turning the Next Gen Development Study into clear guidance for families and educators. First, the student will become familiar with the survey instrument and codebook, reviewing measures of identity, commitment, governance exposure, educational experiences, and intentions to contribute. They will assist with data cleaning, variable construction (e.g., indices of “readiness,” “governance exposure,” or “education intensity”), and documentation to ensure analytical transparency.

The student will then conduct quantitative analyses to identify which combinations of experiences and structures are associated with stronger next gen readiness and intention to engage with family enterprises. This will involve descriptive statistics, cross group comparisons (e.g., students from family business vs. non family backgrounds), and regression or related techniques. Where appropriate, the intern may also conduct basic qualitative coding of open ended responses, organizing themes around challenges, motivations, and perceived support needs.
A major focus is translating findings into accessible, practice oriented outputs. The intern will help build best practice frameworks (e.g., “key elements of an effective next gen development journey”), prepare executive summaries and infographics, and co author a practitioner or business family oriented white paper. Under the mentorship of Dr. Francesco Barbera and the Family Business Institute team in downtown Toronto, the student will receive guidance on both methodological rigor and effective communication.

There are strong opportunities for visibility: the intern’s profile can be featured on the FBI website, and their contribution may be acknowledged on white papers, briefs, or public presentations. The role is particularly valuable for students from family business backgrounds or those aspiring to careers in consulting, HR/organizational development, or family enterprise advisory, offering immediately applicable insights and a standout research experience on their CV.

Skills required:
We seek a student from business, psychology, sociology, education, or related social sciences with strong interest in family enterprise, leadership development, or HR/organizational development. Coursework in statistics/quantitative methods is essential, along with experience using at least one analytical tool (e.g., SPSS, R, Stata, or Python). Familiarity with survey data, scale construction, and basic regression or group comparisons is important. Experience or interest in analyzing open ended responses (e.g., basic qualitative coding) and communicating findings to non academic audiences is highly valued. Curiosity about next gen careers, careful attention to detail, and a collaborative, reflective mindset are critical. Family business experience

3. Best Practices for Thriving Family Businesses in the Greater Toronto Area

This project identifies concrete, evidence-based “best practices” that help family businesses thrive across generations in the Greater Toronto Area (GTA)—one of the world’s most diverse metropolitan regions. Leveraging the Pulse of the GTA Family Business Study, an REB approved dataset of 800+ in depth surveys with family business leaders, we will translate statistical findings into practical, public-facing recommendations for business families, advisors, and policymakers. The dataset captures governance mechanisms (e.g., advisory boards, family councils), the presence and quality of succession plans, perceived advantages and challenges of being a family firm, and GTA specific attributes. We aim to pinpoint patterns associated with positive outcomes such as stronger family–business harmony, clearer leadership pipelines, and greater continuity readiness. Analytically, we will combine descriptive profiling with regression and subgroup comparisons to isolate practices linked to better outcomes, complemented by visual storytelling. The internship is hosted at the Family Business Institute (FBI) at Toronto Metropolitan University’s Ted Rogers School of Management, in the heart of downtown Toronto. Guided by acclaimed academic director Dr. Francesco Barbera, recipient of multiple international research awards, the student will join a vibrant research and practice community. A key emphasis is knowledge mobilization. The student will help craft short best practice briefs, visual summaries (e.g., infographics quantifying the gap between high successor desire and low formal planning), and practitioner white papers or presentations. Contributions can be publicly recognized, including potential profiling on the FBI website and acknowledgment on outputs. For students from family business backgrounds or those pursuing consulting, economic development, or entrepreneurship, the project offers a deep toolkit of evidence based practices and an exceptional, CV worthy applied research experience.

Research area, student roles & skills

Research area: We study evidence-based best practices in family enterprises, focusing on how governance, succession planning, and family–business alignment drive long-term continuity in diverse urban contexts. Centred on the Greater Toronto Area, our work integrates family business scholarship with organizational behavior, entrepreneurship, and migration studies to translate data into practical recommendations. Using large-scale survey evidence, we identify which structures (e.g., boards, councils), processes (e.g., formal plans), and relationship dynamics (e.g., harmony, communication) correlate with positive outcomes across generations. The program is embedded in a leading downtown Toronto research hub dedicated to actionable insights for business families, advisors, and newcomer-founded firms.

Student roles:
The Globalink intern will help transform the Pulse of the GTA Family Business Study into clear, actionable best practices. First, the student will review the survey instrument and codebook, define key outcomes (e.g., harmony, succession readiness), and construct composite indicators for governance and planning quality. They will lead data preparation (cleaning, labeling, handling missingness), exploratory analysis, and documentation to ensure reproducibility.
Next, the student will conduct quantitative analyses: descriptive comparisons of governance and planning patterns, cross tabs by generation/immigrant status, and regressions linking practices to positive outcomes. Where relevant, they will test robustness (e.g., alternative specifications, subgroup analysis) and explore pattern finding (e.g., clustering common practice bundles). They will collaborate with the team and Dr. Francesco Barbera to interpret results in a way that resonates with both academics and practitioners.
A central responsibility is knowledge mobilization. The student will draft concise best practice briefs, build figures/infographics for public reports, and help prepare slide decks or white papers for community presentations. Opportunities for public recognition include potential profiling on the Family Business Institute website and acknowledgment on public facing outputs.
Embedded in a dynamic institute in downtown Toronto, the intern will gain advanced skills in data analysis, visualization, and translation of evidence for non academic audiences—ideal preparation for careers in consulting, economic development, or advising business families. Students from family business backgrounds will gain a practical toolkit directly applicable to their own firms or future client engagements, culminating in a standout experience on their CV.

Skills required:
We seek a student from business, economics, public policy, sociology, or related fields, with strong interest in family enterprise, entrepreneurship, or applied consulting. Essential: coursework in statistics/quantitative methods and proficiency with at least one tool (R, Stata, SPSS, or Python). Experience working with survey data, cleaning/recoding variables, and interpreting regression or subgroup results is important. Strength in communicating findings to non academic audiences—especially data visualization (e.g., ggplot, matplotlib, or Tableau/Power BI)—is highly valued. Curiosity, attention to detail, and teamwork are critical. Familiarity with family business topics or lived experience in a family firm is a strong asset.

4. CEO attributes and corporate decisions

This research project examines CEO characteristics and the Top Management Team's corporate accounting and financial decisions. It builds on the extant psychology research about personality and the growing research in financial economics, corporate finance, accounting and management about TMT characteristics and corporate decisions, and uses data analytics and machine learning methods to study the impact of different CEO individual features on accounting and financial decisions.

Research area, student roles & skills

Research area: My past and current research projects have covered several topics among the most pressing, challenging and impactful research areas in accounting and business administration, namely: corporate ethics and sustainability, corporate transparency, quality of financial reporting, board of directors, executive compensation, auditing, application of big data analytics methods in accounting and crisis management and resilience building during the COVID-19 outbreak. I used a number of qualitative and quantitative methods such as interviews, surveys, archival methods and visual and textual analytics using machine learning techniques.

Student roles:
Search and summarize related previous literature
Collect, clean and prepare data for analysis
Develop appropriate empirical models and run data analyses (descriptive and advanced)
Draw main conclusions and prepare final report

Skills required:
- Data analytics skills
- Coding skills (Python)
- Data cleaning skills
- Ability to work with machine learning models
- Excellent reading and writing english skills

5. ChatbotLLM Enhancements - Supporting REST API, Model Context Protocol, and Learning Tools Interoperability

The project is based on existing research, see https://maiga.athabascau.ca/#chatbotllm, VIP Research Group (https://www.youtube.com/@vipresearchgroup) has: - ChatbotLLM (https://chatbot.vipresearch.ca), is IEEE Northern Canada Section’s Capstone Project Award winnder project, is a system to automatically train chatbots on the materials uploaded by users. ChatbotLLM is not only capable of learning from material in traditional formats like PDF, PPT(X), DOC(X), and TXT, but also can reading XLS(X), image-scanned based PDF, images in PDF, DOC(X), and PPTX, and even MP3 with the help of Google’s Speech Recognition library. Reference: Supun De Silva and Maiga Chang. (2025). ChatbotLLM – Training Educational Chatbots on the Materials Uploaded by Teachers. In: 13th International Conference on Information and Education Technology (ICIET 2025), Fukuyama, Japan, April 18-20, 2025. DOI: 10.1109/ICIET66371.2025.11046253 (IEEE) Access: https://maiga.athabascau.ca/publication/Conference-2025-ICIET2025-ChatbotLLM.pdf The enhancements include but not limited to: 1. Improving as well as simplifying the existing bridge component that includes client and server part. 2. Implementing REST API interfaces similar to OpenAI, Google, and Meta. 3. Implementing Model Context Protocol (MCP) based on the existing bridge component ChatbotLLM has so ChatbotLLM can be acting as Host, Client and Server. 4. Implementing Learning Tools Interoperability (LTI) based on the existing bridge component ChatbotLLM has so ChatbotLLM can be acting as both provider and consumer -- educational platforms like Moodle and other Learning Management Systems (Consumers/Providers) and third-party applications (Providers) can work with ChatbotLLM directly. 5. Implementing the last-mile privacy protection, stateless, and on-the-fly mechanism so the encrypted ans secured content delivered to the existing bridge librarian could maintain its inaccessible by anyone other than ChatbotLLM for the last mile (i.e., training and response generation).

Research area, student roles & skills

Research area: Three research directions Dr. Chang is working on are 1. Game-based Learning and Assessment (i.e., Stealth Assessment): have designed and developed various educational games on different platforms include desktop computers, smartphones, and televisions to assist users learning history and culture, programming language, financial concepts, and help Autism children on activities of daily living as well as elders with disabilities. 2. Data Analytics Service: focus on the design of algorithms and methodologies for learning activity generation, behaviour pattern extraction, and recommendation services. 3. Natural Language Processing: have been appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang).

Student roles:
The detailed function requirements will be given later when the students come to Canada. Based on interview (if there is any) discussions, the student may work on related Generative AI and Natural Language Processing based research (see https://maiga.athabascau.ca/#outcome) or other research project tasks that the student's skill set is fitting better.

The student needs to use browser-side JavaScript, CSS, HTML, and server-side PHP (the use of server-side JavaScript like NodeJS and ExpressJS is **not** allowed). Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps to access. Any functions that requires to use Python scripts beed to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

The student will be able to learn practical modern Python-enabled/supported web system architecture and implementation based on
• front-end browser-side JavaScript, CSS and HTML;
• making AJAX calls to access specific PHP endpoints (i.e., PHP pages);
• bridging by PHP and MariaDB;
• backend service in Python (with or without MariaDB) executed by and respond to PHP calls; and,
• data packaging with JSON for the communications among three layers.

A pilot study is needed to verify the effectiveness of the proposed method and the usability of the app; therefore, the student can know how to design a practical experiment and what statistical research methods can be applied to such research.

Skills required:
For doing this project, the student is better to have the following skills or knowledge:
• [Required] Experience on developing web app with PHP, JSON, HTML, CSS, and AJAX calls.
• [Required] Experience on developing independent application in Python.
• Familiar with Git, MariaDB and SQL commands.
• [Asset] Experience on executing Python script within PHP.
• [Asset] Experience of developing Network Socket and/or WebSocket application in Python.
• [Asset] Capable of adopting Responsive Web Design in web UI design.
• [Asset] Experience of implementing REST API, Model Context Protocol (MCP), and Learning Tools Interoperability (LTI).

6. Climate Governance in the Canadian Corporate Landscape: Commitments, Disclosure, and Accountability

This project investigates how Canadian publicly listed companies are responding to growing climate governance expectations, including regulatory disclosure requirements, investor pressure, and voluntary commitments. Interns will analyze corporate climate disclosures, board-level governance structures, and the gap between stated commitments and measurable action, with the goal of building a comparative picture across sectors.

Research area, student roles & skills

Research area: My research sits at the intersection of corporate governance, sustainable finance, and climate policy. I examine how companies, investors, and regulators are navigating the transition to a low-carbon economy, with a focus on the governance mechanisms — at the board, firm, and market level — that shape corporate climate commitments and accountability. A particular interest is how Canadian companies are responding to evolving domestic and international climate disclosure and governance expectations.

Student roles:
Interns will support the systematic collection and analysis of data on anti-ESG shareholder campaigns, including proxy filings, voting outcomes, and activist communications. They will contribute to literature reviews, help build and maintain a research dataset, and assist in identifying patterns across companies, sectors, and time periods. Interns are expected to work collaboratively as a team and present findings regularly to the supervising researcher.

Skills required:
Familiarity with corporate governance concepts and capital markets
Strong research and analytical skills, including ability to synthesize academic and grey literature
Experience with qualitative or quantitative data collection (e.g., content analysis, database searches)
Comfort working with financial or governance databases (e.g., Bloomberg, ISS, SEC/EDGAR filings) is an asset

7. Corporate misbehavior and CEO career

This research project examines the impact of corporate misconduct on CEO career. Corporate misconduct involves many incidents such as financial scandals, non ethical corporate behavior and environmental incidents. This project aims to study the different consequences for a CEO's career after the announcement of the company's inappropriate behavior.

Research area, student roles & skills

Research area: My past and current research projects have covered several topics among the most pressing, challenging and impactful research areas in accounting and business administration, namely: corporate ethics and sustainability, corporate transparency, quality of financial reporting, board of directors, executive compensation, auditing, application of big data analytics methods in accounting and crisis management and resilience building during the COVID-19 outbreak. I used a number of qualitative and quantitative methods such as interviews, surveys, archival methods and visual and textual analytics using machine learning techniques.

Student roles:
The research assistant is expected to:
- Search, read and summarize relevant published papers,
- Collect, clean and analyze data using different tools such as Excel, Stata
- Write research reports

Skills required:
- Accounting or Computer Science or Statistics and Econometrics with some coding
skills using Stata, R, Python
- Excellent reading and writing skills in English.

8. Cross Cultural Patterns in Next Gen Identity, Intentions, and Governance Exposure

This project investigates how culture and family migration history shape next gen identity, intentions, and governance exposure in family business contexts, leveraging Toronto’s diversity as a natural laboratory. Using the Family Business Institute’s Research Ethics Board approved 900 survey Next-Gen dataset from the Ted Rogers School student pool, we analyze cross cultural variation and common ground across key domains: demographics and family business background; governance structures and practices; identity/commitment; intentions to return in operating or ownership roles; family business education; and open ended narratives on challenges and opportunities. Analytically, the study will combine descriptive profiling, segmentation by cultural background and newcomer status, and regression models with interaction terms to test whether education, mentoring, and governance exposure relate differently to readiness and intentions across groups. We will complement the quantitative work with systematic coding of open ended responses to surface culturally nuanced themes around obligation, autonomy, communication, and perceived career fit. Where informative, we will develop comparative profiles and typologies that illustrate inclusive pathways that resonate across cultures. Hosted at the Family Business Institute at Toronto Metropolitan University’ in the heart of downtown Toronto, the internship embeds the student in a vibrant hub that bridges scholarship and practice. Under the mentorship of award-winning scholar Dr. Francesco Barbera, the student will gain rigorous analytical training and experience in knowledge mobilization. Outputs will include data visualizations, comparative dashboards, and a research to practice brief highlighting inclusive development approaches for owners, advisors, and educators. Recognition opportunities include a profile on the Institute’s website and name credit on white papers or public presentations. For students from family business backgrounds, or those aspiring to work with diverse business families, this project offers a nuanced understanding of how culture intersects with education and governance exposure, building a toolkit directly applicable to advisory, policy, and leadership development.

Research area, student roles & skills

Research area: Our specialized research area examines cross cultural patterns in next generation development within family enterprises. We integrate family business, migration studies, and organizational behavior to analyze how cultural background and family migration history shape identity/commitment, intentions to join or own the business, and exposure to governance practices (e.g., boards, councils, structured meetings). Grounded in Toronto’s unparalleled diversity and using a 900+ respondent, ethics approved student dataset, we quantify differences and commonalities across cultural segments and newcomer status. The program, based at the Family Business Institute in downtown Toronto, translates rigorous evidence into inclusive, practical guidance for owners, advisors, and educators.

Student roles:
The Globalink intern will lead analyses that map cross cultural patterns in next gen identity, intentions, and governance exposure. First, the student will review the dataset; harmonize cultural background and newcomer status variables; and document coding decisions to ensure transparency. They will clean data and construct indices for identity/commitment, readiness, governance exposure, and education intensity.

Next, the intern will conduct descriptive profiling and segmentation by cultural background and migration history, followed by regression models with interaction terms to test whether education, mentoring, and governance exposure predict intentions and readiness differently across groups. Robustness checks may include alternative specifications, stratified models, and sensitivity analyses. The student will also code open-ended responses to surface themes such as obligation, intergenerational communication, perceived barriers, and support needs, integrating qualitative insights with quantitative findings.

Knowledge mobilization is central. The intern will create comparative profiles and visualizations (e.g., heatmaps, small multiples, or dashboards) and co-author a research to practice brief outlining inclusive approaches that resonate across cultures. They will help prepare slide decks and concise summaries for multiple audiences, ensuring clarity for non-specialist audiences.

Based at the Family Business Institute in downtown Toronto and mentored by award winning scholar Dr. Francesco Barbera, the student will develop advanced mixed methods skills and evidence translation capabilities within a vibrant hub that bridges research and practice. Visibility opportunities include a profile on the Institute’s website and name credit on white papers or public presentations. Regular research check ins will reinforce rigour and professional growth.

The role is especially meaningful for students from family businesses or those aiming to work with diverse business families. By the end of the internship, the student will have a nuanced, data driven understanding of how culture intersects with education and governance exposure, plus a portfolio of public-facing outputs and a standout CV-ready experience.

Skills required:
We seek a student from business, psychology, sociology, economics, or education with interest in family enterprise, migration, or cross cultural management. Essential: coursework in statistics/quantitative methods and proficiency with at least one tool (R, Stata, SPSS, or Python). Experience preparing survey data, constructing indices, and running regressions or group comparisons is important. Added value: sensitivity to cross cultural research, basic qualitative coding of open ended text, and skill in data visualization (ggplot, matplotlib, Tableau, or Power BI). Strong writing for non academic audiences, curiosity, and collaboration are critical. Lived experience in multicultural settings or family business is a plus.

9. Developing an AI-Enhanced Confidence-Based Adaptive Practicing (CAP) System for Online Education

This project focuses on designing and developing the Confidence-Based Adaptive Practicing (CAP) system, an AI-enhanced adaptive learning platform. The CAP system is grounded in learning science, specifically utilizing the Zone of Proximal Development-informed Knowledge Tracing (ZPD-KT) model. It aims to address "learning fog," which is the persistent uncertainty online learners face regarding what they have mastered, where their knowledge gaps lie, and their exam readiness. The overarching goal of the research is to build a full prototype featuring a front-end adaptive engine and a Large Language Model (LLM) service layer. This system is designed to provide dialogic conversational scaffolding, on-demand micro-lessons, and graduated hints through natural language. The completed prototype will be piloted in university-level STEM courses to analyze whether LLM-assisted practice reduces learner "wheel-spinning" and improves mastery progression. The project would be co-supervised by Dr. Hongxin YAN and Dr. Maiga CHANG.

Research area, student roles & skills

Research area: Three research directions Dr. Chang is working on are 1. Game-based Learning and Assessment (i.e., Stealth Assessment): have designed and developed various educational games on different platforms include desktop computers, smartphones, and televisions to assist users learning history and culture, programming language, financial concepts, and help Autism children on activities of daily living as well as elders with disabilities. 2. Data Analytics Service: focus on the design of algorithms and methodologies for learning activity generation, behaviour pattern extraction, and recommendation services. 3. Natural Language Processing: have been appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang).

Student roles:
The student will be instrumental in the technical development of the CAP system prototype during their summer internship. Below are the tasks that the student can choose (one or more):
• Task-1: Building out the CAP front-end web application. This involves programming user interfaces for a student mastery dashboard and implementing visual wheel-spinning alerts.
• Task-2: The student will heavily contribute to building the LLM Service Layer of the platform. Specific front-end tasks include developing an LLM Chat Panel component that features pre-answer and post-answer interaction modes, as well as quick-action buttons for the learners.
• Task-3: The role will involve assisting with the engineering of the backend architecture, including the Prompt Orchestrator, Context Manager, and Streaming Response Handler to ensure seamless LLM integration.
• Task-4: The student will help establish the data logging mechanisms required to capture anonymized learning trajectories, LLM interaction logs, and practice redirection events to support the project's future learning analytics research.

The student needs to use browser-side JavaScript, CSS, HTML, and server-side PHP (the use of server-side JavaScript like NodeJS and ExpressJS is **not** allowed). Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps to access. Any functions that requires to use Python scripts need to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

Skills required:
The ideal candidate should possess strong software development skills, particularly in front-end web development. The student should also have programming experience with Python, which will be utilized for backend analytics pipelines. Prior experience with API integration, prompt engineering, or working directly with Large Language Model (LLM) services is a strong asset.
• [Required] Experience on developing web-app with PHP, JSON, HTML, CSS, and AJAX calls.
• [Required] Experience on developing independent Python applications.
• [Asset] Experience on executing Python script within PHP.
• [Asset] Having experience on using Python to implement Model Context Protocol (MCP) and Learning Tools Interoperability (LTI).

10. Driving and Walking Navigation Android App built on Open Street Map and Next-Stop Recommender

The project is enhancing the following existing research VIP Research Group (https://www.youtube.com/@vipresearchgroup) has: - Next Stop Recommender (https://nextstop.vipresearch.ca/) which can provide users recommendation for their next visit according to the similarity of their behaviours between each others and the connections amongst locations. - he Generative AI (GenAI) supported Note Takers App (https://notetakers.vipresearch.ca) is an innovative tool designed to help users capture and organize notes from spoken conversations, incorporates Public Key Infrasture (PKI), Secure Sockets Layer (SSL) and salted secured hash algorithm. The research project involves three parts: (1) The production server needs to (a) have Open Street Map Tile Server and Routing Server for selected countries and regions; (b) offers endpoints that allows client apps (including the Android App) to download selected region's map; (c) receive the requests for synchronizing users their location bookmarks and app's configuration anonymously in a secure and privacy protected way; (d) allow administrators to test new Text-Generation, Speech-to-Text, and Text-to-Speech (TTS) GGUF models and auto update available model list and corresponding voice options (for TTS models); (e) offers endpont for accessing server-side services on Text-Generation and Speech-to-Text. (2) The Android App allows users to (a) download an available Text-Generation, Speech-to-Text, and Text-to-Speech (TTS) GGUF model from the production server according to the available model list provided by the endpoint; (b) choose pre-downloaded models and a supported voice from the selected TTS model; (c) ask the app to find specific place or address or coordinates for navigation via talking to the app with natural language; (d) listen to the app for navigation and interactions; (e) use the endpoint the production server offers to synchronize device uuid, corresponding configurations, and bookmarks anonymously in a secure and privacy protected way. The app can take NoteTakers' secure communication workflow and design as reference (see https://maiga.athabascau.ca/#notetakers and https://notetakers.vipresearch.ca).

Research area, student roles & skills

Research area: Three research directions Dr. Chang is working on are 1. Game-based Learning and Assessment (i.e., Stealth Assessment): have designed and developed various educational games on different platforms include desktop computers, smartphones, and televisions to assist users learning history and culture, programming language, financial concepts, and help Autism children on activities of daily living as well as elders with disabilities. 2. Data Analytics Service: focus on the design of algorithms and methodologies for learning activity generation, behaviour pattern extraction, and recommendation services. 3. Natural Language Processing: have been appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang).

Student roles:
The detailed function requirements will be given later when the students come to Canada. Based on interview (if there is any) discussions, the student may work on related Generative AI and Natural Language Processing based research (see https://maiga.athabascau.ca/#outcome) or other research project tasks that the student's skill set is fitting better.

The student needs to use Android Studio with either Kotlin or Java to develop the mobile app as well as follow Google Play's policy and requirements to prepare necessary materials for publishing and updating the app. The student needs to fix the issues brought up by Google Play while publishing the app for VIP Research Group. Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps (including the mobile app) to access. Any functions that requires to use Python scripts beed to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

The student will be able to learn practical modern Python-enabled/supported web system architecture and implementation based on
• front-end browser-side JavaScript, CSS and HTML;
• making AJAX calls to access specific PHP endpoints (i.e., PHP pages);
• bridging by PHP and MariaDB;
• backend service in Python (with or without MariaDB) executed by and respond to PHP calls; and,
• data packaging with JSON for the communications among three layers.

A pilot study is needed to verify the effectiveness of the proposed method and the usability of the app; therefore, the student can know how to design a practical experiment and what statistical research methods can be applied to such research.

Skills required:
For doing this project, the student is better to have the following skills or knowledge:
• [Required] Experience on developing app with Android in either Kotlin or Java.
• [Required] Experience on developing web app with PHP, JSON, HTML, CSS, and AJAX calls.
• [Required] Experience on developing independent application in Python.
• Familiar with Git, mySQL/MariaDB and SQL commands.
• [Asset] Experience on executing Python script within PHP.
• [Asset] Experience on installing Open Street Map Tile Server and Routing Server on Ubuntu 24.04.
• [Asset] Experience in using Text-Generation, Speech-to-Text, and Text-to-Speech (TTS) GGUF models in Python.

11. Generative AI supported Next Stop Recommender Android App

The project is enhancing the following existing research VIP Research Group (https://www.youtube.com/@vipresearchgroup) has: - Next Stop Recommender (https://nextstop.vipresearch.ca/) which can provide users recommendation for their next visit according to the similarity of their behaviours between each others and the connections amongst locations. - he Generative AI (GenAI) supported Note Takers App (https://notetakers.vipresearch.ca) is an innovative tool designed to help users capture and organize notes from spoken conversations, incorporates Public Key Infrasture (PKI), Secure Sockets Layer (SSL) and salted secured hash algorithm. The research project involves three parts: (1) The production server needs to (a) have Open Street Map Tile Server and Routing Server for selected countries and regions; (b) offers endpoints that allows client apps (including the Android App) to download selected region's map; (c) receive the requests for synchronizing users their historical records, preferences and app's configuration anonymously in a secure and privacy protected way; (d) allow administrators to test new Text-Generation, Speech-to-Text, and Text-to-Speech (TTS) GGUF models and auto update available model list and corresponding voice options (for TTS models); (e) offers endpont for accessing server-side services on Text-Generation and Speech-to-Text. (2) The Android App allows users to (a) download an available Text-Generation, Speech-to-Text, and Text-to-Speech (TTS) GGUF model from the production server according to the available model list provided by the endpoint; (b) choose pre-downloaded models and a supported voice from the selected TTS model; (c) ask the app to recommend places to visit for next via talking to the app with natural language; (d) listen to the app to explain the recommendations; (e) use the endpoint the production server offers to synchronize device uuid, corresponding configurations, and historical records anonymously in a secure and privacy protected way. The app can take NoteTakers' secure communication workflow and design as reference (see https://maiga.athabascau.ca/#notetakers and https://notetakers.vipresearch.ca).

Research area, student roles & skills

Research area: Three research directions Dr. Chang is working on are 1. Game-based Learning and Assessment (i.e., Stealth Assessment): have designed and developed various educational games on different platforms include desktop computers, smartphones, and televisions to assist users learning history and culture, programming language, financial concepts, and help Autism children on activities of daily living as well as elders with disabilities. 2. Data Analytics Service: focus on the design of algorithms and methodologies for learning activity generation, behaviour pattern extraction, and recommendation services. 3. Natural Language Processing: have been appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang).

Student roles:
The detailed function requirements will be given later when the students come to Canada. Based on interview (if there is any) discussions, the student may work on related Generative AI and Natural Language Processing based research (see https://maiga.athabascau.ca/#outcome) or other research project tasks that the student's skill set is fitting better.

The student needs to use Android Studio with either Kotlin or Java to develop the mobile app as well as follow Google Play's policy and requirements to prepare necessary materials for publishing and updating the app. The student needs to fix the issues brought up by Google Play while publishing the app for VIP Research Group. Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps (including the mobile app) to access. Any functions that requires to use Python scripts beed to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

The student will be able to learn practical modern Python-enabled/supported web system architecture and implementation based on
• front-end browser-side JavaScript, CSS and HTML;
• making AJAX calls to access specific PHP endpoints (i.e., PHP pages);
• bridging by PHP and MariaDB;
• backend service in Python (with or without MariaDB) executed by and respond to PHP calls; and,
• data packaging with JSON for the communications among three layers.

A pilot study is needed to verify the effectiveness of the proposed method and the usability of the app; therefore, the student can know how to design a practical experiment and what statistical research methods can be applied to such research.

Skills required:
For doing this project, the student is better to have the following skills or knowledge:
• [Required] Experience on developing app with Android in either Kotlin or Java.
• [Required] Experience on developing web app with PHP, JSON, HTML, CSS, and AJAX calls.
• [Required] Experience on developing independent application in Python.
• Familiar with Git, mySQL/MariaDB and SQL commands.
• [Asset] Experience on executing Python script within PHP.
• [Asset] Experience on installing Open Street Map Tile Server and Routing Server on Ubuntu 24.04.
• [Asset] Experience in using Text-Generation, Speech-to-Text, and Text-to-Speech (TTS) GGUF models in Python.

12. Generative AI supported Visualized Information Personal (VIP) Reading/Reader Android App

The research project involves three parts: (1) The production server that offer endpoints for client apps (including the Android App) to (a) receive the requests for synchronizing users their reading progress, annotations, and bookmarks anonymously in a secure and privacy protected way; (b) allow administrators to test new Text-to-Speech (TTS) GGUF models and auto update available model list and corresponding voice options. (2) The Android App allows users to (a) choose a folder to scan for documents (include but not limited to PDF, EPUB, DOCX, TXT, etc.) to create a library list for reading and reader; (b) select a document from any folder to automatically copy and save into the scan folder and update the library list; (c) enable the background scan process so when the app is open it will auto scan and update the library list; (d) read, annotate, and bookmark a document in the way in line with modern e-book readers use; (e) use the endpoint the production server offers to synchronize document hash values (i.e., unique ids), corresponding reading progress, annotations, and bookmarks anonymously in a secure and privacy protected way. The app can take NoteTakers' secure communication workflow and design as reference (see https://maiga.athabascau.ca/#notetakers and https://notetakers.vipresearch.ca). (3) The Android App allows users to (a) download an available Text-to-Speech (TTS) GGUF model from the production server according to the available model list provided by the endpoint; (b) choose a pre-downloaded model and a supported voice for reader; (c) ask the app to read (as well as pause and stop reading) the document content according to the content's language; (d) synchronize the reader's progress.

Research area, student roles & skills

Research area: Dr. Chang are working on continuously on three directions: AI and Chatbot, Data Analytics Service, Mobile App and Game Development. Dr. Chang was appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang) working on Natural Language Processing related research that can automatically summarize the text the machine read and learnt for users' questions (see http://maiga.athabascau.ca/#askcovidq and https://ask4summary.vipresearch.ca), IEEE Northern Canada Section’s Capstone Project Award winnder project, ChatbotLLM (https://chatbot.vipresearch.ca/) that train chatbots on the materials uploaded.

Student roles:
The detailed function requirements will be given later when the students come to Canada. Based on interview (if there is any) discussions, the student may work on related Generative AI and Natural Language Processing based research (see https://maiga.athabascau.ca/#outcome) or other research project tasks that the student's skill set is fitting better.

The student needs to use Android Studio with either Kotlin or Java to develop the mobile app as well as follow Google Play's policy and requirements to prepare necessary materials for publishing and updating the app. The student needs to fix the issues brought up by Google Play while publishing the app for VIP Research Group. Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps (including the mobile app) to access. Any functions that requires to use Python scripts beed to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

The student will be able to learn practical modern Python-enabled/supported web system architecture and implementation based on
• front-end browser-side JavaScript, CSS and HTML;
• making AJAX calls to access specific PHP endpoints (i.e., PHP pages);
• bridging by PHP and MariaDB;
• backend service in Python (with or without MariaDB) executed by and respond to PHP calls; and,
• data packaging with JSON for the communications among three layers.

A pilot study is needed to verify the effectiveness of the proposed method and the usability of the app; therefore, the student can know how to design a practical experiment and what statistical research methods can be applied to such research.

Skills required:
For doing this project, the student is better to have the following skills or knowledge:
• [Required] Having experience in developing app with Android in either Kotlin or Java.
• [Required] Having experience in developing web app with PHP, JSON, HTML, CSS, and AJAX calls.
• [Required] Having experience in developing independent application in Python.
• Familiar with Git commands.
• Familiar with MariaDB and SQL commands.
• [Asset] Capable of adopting Responsive Web Design in web UI design.
• [Asset] Having experience in using Text-to-Speech (TTS) GGUF models in Python.

13. Generative AI supported Visualized Information Platform (VIP) for research analysis

The project is based on existing research, see https://maiga.athabascau.ca/#chatbotllm, VIP Research Group (https://www.youtube.com/@vipresearchgroup) has: - ChatbotLLM (https://chatbot.vipresearch.ca), is IEEE Northern Canada Section’s Capstone Project Award winnder project, is a system to automatically train chatbots on the materials uploaded by users. ChatbotLLM is not only capable of learning from material in traditional formats like PDF, PPT(X), DOC(X), and TXT, but also can reading XLS(X), image-scanned based PDF, images in PDF, DOC(X), and PPTX, and even MP3 with the help of Google’s Speech Recognition library. Reference: Supun De Silva and Maiga Chang. (2025). ChatbotLLM – Training Educational Chatbots on the Materials Uploaded by Teachers. In: 13th International Conference on Information and Education Technology (ICIET 2025), Fukuyama, Japan, April 18-20, 2025. DOI: 10.1109/ICIET66371.2025.11046253 (IEEE) Access: https://maiga.athabascau.ca/publication/Conference-2025-ICIET2025-ChatbotLLM.pdf - Mobile Accessible Seamless Agent (MASA) (https://masa.chatbot.vipresearch.ca), removes authentication barriers such as complex logins ensures instant access to learning materials on students’ devices via a familiar, intuitive interface. MASA’s UI/UX follows common mobile app patterns to minimize cognitive load and support natural, companion-like interaction between learners and the Chatbot service provider. Reference: Fuzheng ZHAO, Zibusiso MAFAITI, Yongjin LI, and Maiga CHANG. (2026). The research project aims to utilize the content uploaded, processed, and learned by ChatbotLLM to incorporate browser-side JavaScript libraries (server-side JavaScript like NodeJS and ExpressJS will not be considered) for visualizing dynamic and automatic relations among data via the user interactions (e.g., clicking, double clicking, typing, selecting, chatting, etc.). This project wants to have the functions supported by existing AI tools like visualization and interactivity functions in Research Rabbit (https://www.researchrabbit.ai/features), research search and summarization functions in Vectara (https://www.vectara.com/developers/build/demos), chatbot like research search and summarization functions in GPT-Trainer (https://gpt-trainer.com/technology), and Elsevier's AI Generated Narrative for Researchers function (https://www.elsevier.com/academic-and-government/how-elsevier-helps-funders-build-research-impact-stories).

Research area, student roles & skills

Research area: Three research directions Dr. Chang is working on are 1. Game-based Learning and Assessment (i.e., Stealth Assessment): have designed and developed various educational games on different platforms include desktop computers, smartphones, and televisions to assist users learning history and culture, programming language, financial concepts, and help Autism children on activities of daily living as well as elders with disabilities. 2. Data Analytics Service: focus on the design of algorithms and methodologies for learning activity generation, behaviour pattern extraction, and recommendation services. 3. Natural Language Processing: have been appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang).

Student roles:
The detailed function requirements will be given later when the students come to Canada. Based on interview (if there is any) discussions, the student may work on related Generative AI and Natural Language Processing based research (see https://maiga.athabascau.ca/#outcome) or other research project tasks that the student's skill set is fitting better.

The student needs to use browser-side JavaScript, CSS, HTML, and server-side PHP (the use of server-side JavaScript like NodeJS and ExpressJS is **not** allowed). Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps to access. Any functions that requires to use Python scripts beed to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

The student will be able to learn practical modern Python-enabled/supported web system architecture and implementation based on
• front-end browser-side JavaScript, CSS and HTML;
• making AJAX calls to access specific PHP endpoints (i.e., PHP pages);
• bridging by PHP and MariaDB;
• backend service in Python (with or without MariaDB) executed by and respond to PHP calls; and,
• data packaging with JSON for the communications among three layers.

A pilot study is needed to verify the effectiveness of the proposed method and the usability of the app; therefore, the student can know how to design a practical experiment and what statistical research methods can be applied to such research.

Skills required:
For doing this project, the student is better to have the following skills or knowledge:
• [Required] Experience on developing web app with PHP, JSON, HTML, CSS, and AJAX calls.
• [Required] Experience on developing independent application in Python.
• [Required] Experience of using Chart.js, D3.js or alternative solutions.
• [Required] Extracting important data into JSON with Large Language Model with Python.
• Familiar with Git, MariaDB and SQL commands.
• [Asset] Capable of adopting Responsive Web Design in web UI design.
• [Asset] Experience on executing Python script within PHP.

14. Governance Structures and Their Impact in Toronto Family Businesses

This project investigates how governance structures influence performance, succession planning, and long term continuity in GTA family firms. While many families adopt boards, advisory councils, family councils, or written policies, evidence on which mechanisms truly support effective transitions and harmony remains fragmented. We address this gap using the Pulse of the GTA Family Business Study—an REB approved dataset with 800+ surveys capturing governance arrangements, succession intentions and plans, firm size and generation, and key demographic and regional characteristics. The core objective is to identify which governance mechanisms and role clarity practices are associated with stronger succession preparedness and positive reported outcomes (e.g., family–business harmony, conflict management, leadership pipeline clarity). The study will combine descriptive statistics, segmentation of firms by governance type, and regression or subgroup comparisons to isolate relationships between governance design and outcomes. Where appropriate, we will develop typologies or “governance profiles” that illustrate effective bundles of practices. Hosted at the Family Business Institute (FBI) at Toronto Metropolitan University’s Ted Rogers School of Management—in the heart of downtown Toronto—the internship embeds the student in a vibrant research and practice community. Under the mentorship of acclaimed academic director Dr. Francesco Barbera, recipient of multiple international research awards, the intern will engage with cutting edge research and a growing practitioner network. A strong emphasis is placed on knowledge mobilization. The student will help craft executive summaries, visual dashboards, and concise best practice notes to inform owners, advisors, and policymakers. Contributions can be publicly recognized through potential profiling on the FBI website and acknowledgment on white papers or presentations. For students with family business experience or aspirations in consulting, finance, or management, this project offers advanced analytical training, deep governance insights, and a standout, CV worthy applied research experience.

Research area, student roles & skills

Research area: We study governance in family enterprises, examining how boards, advisory councils, family councils, role clarity, and formal policies shape decision making, oversight, and intergenerational continuity. Anchored in Toronto's diverse context, our work connects family business scholarship with organizational governance and succession research to identify which structures and processes support performance, planning, and harmony. Using large scale survey evidence from the Pulse of the GTA Family Business Study, we analyze how governance design relates to succession readiness and reported outcomes. The program operates from a leading downtown Toronto hub dedicated to translating rigorous evidence into actionable insights for business families.

Student roles:
The Globalink intern will lead core analytical tasks linking governance to outcomes in GTA family firms. First, the student will review the survey instrument and codebook, inventory governance measures (e.g., boards, advisory councils, family councils, policies, role delineation), and define target outcomes (e.g., succession readiness, harmony, conflict management, leadership pipeline). They will clean and document data, construct composite indices (e.g., governance formality or role clarity scores), and prepare reproducible analysis files.

Next, the student will conduct quantitative analyses: descriptive comparisons across governance types, cross tabs by generation/immigrant status, and regression models testing associations between governance mechanisms and key outcomes, with robustness checks (alternative specifications, subgroup analysis). Where suitable, they will develop governance typologies or profiles (e.g., clustering bundled practices) and produce clear visualizations to communicate patterns.

Knowledge mobilization is central. The intern will draft executive summaries and short best practice briefs, build figures/infographics and interactive or static dashboards for public facing reports, and help prepare slide decks or white papers for community presentations. Mentored by Dr. Francesco Barbera and the FBI team in downtown Toronto, the student will gain advanced skills in empirical analysis, visualization, and translation of evidence for owners, advisors, and policymakers.

There are strong visibility opportunities: the intern may be profiled on the Family Business Institute website and acknowledged on public outputs. The role is especially valuable for students from family businesses or those aiming for consulting, finance, or management careers. By project’s end, the student will possess a practical governance toolkit and a standout research experience on their CV.

Skills required:
Ideal candidates come from business, economics, public policy, or related social sciences, with interest in family enterprise and governance. Essential: coursework in statistics/quantitative methods and proficiency with at least one tool (R, Stata, SPSS, or Python). Experience managing survey data, constructing indices, and interpreting regression or subgroup results is important. Skills in visualization or dashboards (e.g., ggplot, matplotlib, Tableau, or Power BI) and clear writing for non academic audiences are highly valued. Curiosity, rigor, and collaboration are critical. Familiarity with family business topics—or lived experience in a family firm—is a strong asset, especially for translating findings into practical guidance.

15. Key Phrases Extraction based on Next Word Prediction

The project is based on existing research, see http://maiga.athabascau.ca/#askcovidq, VIP Research Group (https://www.youtube.com/@vipresearchgroup) has: - Ask4Summary (https://ask4summary.vipresearch.ca/) processes these text-based content, it uses Natural Language Processing (NLP) techniques that include tokenization, n-grams extraction, and part-of-speech (PoS) tagging. It then identifies the keywords from a user’s question and uses cosine similarity to summarize the associated content and present to the user. For a regular graduate level Academic Writing course, Ask4Summary can complete the reading of the course content in couple of hours and respond user's question in 0.7 seconds. Currently Ask4Summary can serve users via three ways: web system, Moodle plugin, and chatbot. Reference: Rita Kuo, Maria F. Iriarte, Di Zou and Maiga Chang. (2023). Preliminary Performance Assessment on Ask4Summary’s Reading Methods for Summary Generation. In: 19th International Conference on Intelligent Tutoring Systems, (ITS 2023), Hybrid, Corfu, Greece, June 2-June 5, 2023. (Springer) Access: https://maiga.athabascau.ca/publication/Conference-2023-ITS2023-Ask4Summary.pdf The research aims to adopt the concept and approaches of next word prediction and focuses directly on the n-gram relationships. This research will use neural networks for training and obtaining word vectors. The research can validate the potential of a given key phrase (i.e., an n gram word) reversely with the trained model’s outputs, a possible n-gram list, predicted from the (n-1)-gram. For instance, given a 3-gram “the research river” and feeds its 2-gram “the research” into the trained mode, the potential 3-gram list includes “the research project”, “the research goal”, and “the research issue” could be provided the trained model; however, “the research river” is not in the list. In such case “the research river” might be considered as an uncommon word even it has exactly same part-of-speech tag combination (i.e., DT-NN-NN) like other words that starts with “the research” in the 3-gram list.

Research area, student roles & skills

Research area: Three research directions Dr. Chang is working on are 1. Game-based Learning and Assessment (i.e., Stealth Assessment): have designed and developed various educational games on different platforms include desktop computers, smartphones, and televisions to assist users learning history and culture, programming language, financial concepts, and help Autism children on activities of daily living as well as elders with disabilities. 2. Data Analytics Service: focus on the design of algorithms and methodologies for learning activity generation, behaviour pattern extraction, and recommendation services. 3. Natural Language Processing: have been appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang).

Student roles:
The detailed function requirements will be given later when the students come to Canada. Based on interview (if there is any) discussions, the student may work on related Generative AI and Natural Language Processing based research (see https://maiga.athabascau.ca/#outcome) or other research project tasks that the student's skill set is fitting better.

The student needs to use browser-side JavaScript, CSS, HTML, and server-side PHP (the use of server-side JavaScript like NodeJS and ExpressJS is **not** allowed). Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps to access. Any functions that requires to use Python scripts beed to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

The student will be able to learn practical modern Python-enabled/supported web system architecture and implementation based on
• front-end browser-side JavaScript, CSS and HTML;
• making AJAX calls to access specific PHP endpoints (i.e., PHP pages);
• bridging by PHP and MariaDB;
• backend service in Python (with or without MariaDB) executed by and respond to PHP calls; and,
• data packaging with JSON for the communications among three layers.

A pilot study is needed to verify the effectiveness of the proposed method and the usability of the app; therefore, the student can know how to design a practical experiment and what statistical research methods can be applied to such research.

Skills required:
For doing this project, the student is better to have the following skills or knowledge:
• [Required] Experience on developing web app with PHP, JSON, HTML, CSS, and AJAX calls.
• [Required] Experience on developing independent application in Python.
• [Required] Experience of using Natural Language Tool Kit in Python.
• [Required] Experience of training Neural Network Models for text in Python.
• Familiar with Git, MariaDB and SQL commands.
• [Asset] Experience on executing Python script within PHP.
• [Asset] Capable of adopting Responsive Web Design in web UI design.

16. Memorize book reading Android App based on GAIOLA

The research project is based on existing research, see https://maiga.athabascau.ca/#gaiola: - Generative AI supported Oral Language Activity (GAIOLA)(https://gaiola3.vipresearch.ca/) allows students to write the first draft of their script with whatever means they choose to employ, and then helping students practice, rehearse, and iterate on their script with the help of AI. By the end of this process, students will have a script that they can deliver well orally, and they will have practiced delivering it multiple times, so that they can be confident in their oral performance. Reference: Hsieh-Jun CHEN, Yongjin LI, Syed Ali MANSOOR, Han-Chieh CHEN, and Maiga Chang. (2026). GAIOLA: An AI-assisted rehearsal platform for output-oriented EFL oral presentation practice. In: 26th IEEE International Conference on Advanced Learning Technologies (ICALT 2026), Hung Yen, Vietnam, July 6-9, 2026. (IEEE)(accepted) Access: https://maiga.athabascau.ca/publication/Conference-2026-ICALT-GAIOLA.pdf - The Generative AI (GenAI) supported Note Takers App (https://notetakers.vipresearch.ca/) is an innovative tool designed to help users capture and organize notes from spoken conversations. Its incorporates Public Key Infrasture (PKI), Secure Sockets Layer (SSL) and salted secured hash algorithm. The research project involves two parts: (1) the production server that offers endpoints for client apps (including the Android App) to synchronize reading activities and user's practice data in a secure and privacy protection mechanics as well as allows administrators to check individual's progress and performances on a tile-style dashboard with interactable visualized charts (e.g., bar chart, line chart, radar chart, etc.), numbers and statistics; (2) the Android App allows users to switch activities, listen to the computer reading the content, and practice with and without content as hint as well as synchronize practice data via endpoints offered by the production server when the device has Internet access. The app can take NoteTakers' secure communication workflow (https://maiga.athabascau.ca/#notetakers) and design as well as GAIOLA's UI and workflow as reference.

Research area, student roles & skills

Research area: Dr. Chang are working on continuously on three directions: AI and Chatbot, Data Analytics Service, Mobile App and Game Development. Dr. Chang was appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang) working on Natural Language Processing related research that can automatically summarize the text the machine read and learnt for users' questions (see http://maiga.athabascau.ca/#askcovidq and https://ask4summary.vipresearch.ca), IEEE Northern Canada Section’s Capstone Project Award winnder project, ChatbotLLM (https://chatbot.vipresearch.ca/) that train chatbots on the materials uploaded.

Student roles:
The detailed function requirements will be given later when the students come to Canada. Based on interview (if there is any) discussions, the student may work on related Generative AI and Natural Language Processing based research (see https://maiga.athabascau.ca/#outcome) or different research project's tasks according to the student's skill set.

The student needs to use Android Studio with either Kotlin or Java to develop the mobile app as well as follow Google Play's policy and requirements to prepare necessary materials for publishing and updating the app. The student needs to fix the issues brought up by Google Play while publishing the app for VIP Research Group. Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps (including the mobile app) to access. Any functions that requires to use Python scripts beed to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

The student will be able to learn practical modern Python-enabled/supported web system architecture and implementation based on
• front-end browser-side JavaScript, CSS and HTML;
• making AJAX calls to access specific PHP endpoints (i.e., PHP pages);
• bridging by PHP and MariaDB;
• backend service in Python (with or without MariaDB) executed by and respond to PHP calls; and,
• data packaging with JSON for the communications among three layers.

A pilot study is needed to verify the effectiveness of the proposed method and the usability of the app; therefore, the student could know how to design a practical experiment and what statistical research methods can be applied to such research.

Skills required:
For doing this project, the student is better to have the following skills or knowledge:
• [Required] Having experience in developing app with Android in either Kotlin or Java.
• [Required] Having experience in developing web app with PHP, JSON, HTML, CSS, and AJAX calls.
• [Required] Having experience in developing independent application in Python.
• Familiar with Git commands.
• Familiar with MariaDB and SQL commands.
• [Asset] Capable of adopting Responsive Web Design in web UI design.
• [Asset] Having experience in using Natural Language Processing library (NLTK) and LLMs in Python.

17. Mobile Accessible Seamless Agent (MASA) Android App

The project is enhancing the following existing research, see https://maiga.athabascau.ca/#chatbotllm, VIP Research Group (https://www.youtube.com/@vipresearchgroup) has: - ChatbotLLM (https://chatbot.vipresearch.ca), is IEEE Northern Canada Section’s Capstone Project Award winnder project, is a system to automatically train chatbots on the materials uploaded by users. ChatbotLLM is not only capable of learning from material in traditional formats like PDF, PPT(X), DOC(X), and TXT, but also can reading XLS(X), image-scanned based PDF, images in PDF, DOC(X), and PPTX, and even MP3 with the help of Google’s Speech Recognition library. Different from other Generative AI tools or systems, all the responses were identified and generated based on the content users provided instead of generating things based on everything it learned from Internet or other sources and trying to answering everything include hallucinations for you. So ChatbotLLM would tell you "NO, I don't know" if it cannot find any relevant content for your question from the given materials uploaded. Reference: Supun De Silva and Maiga Chang. (2025). ChatbotLLM – Training Educational Chatbots on the Materials Uploaded by Teachers. In: 13th International Conference on Information and Education Technology (ICIET 2025), Fukuyama, Japan, April 18-20, 2025. DOI: 10.1109/ICIET66371.2025.11046253 (IEEE) Access: https://maiga.athabascau.ca/publication/Conference-2025-ICIET2025-ChatbotLLM.pdf - Mobile Accessible Seamless Agent (MASA) (https://masa.chatbot.vipresearch.ca), removes authentication barriers such as complex logins ensures instant access to learning materials on students’ devices via a familiar, intuitive interface. MASA’s UI/UX follows common mobile app patterns to minimize cognitive load and support natural, companion-like interaction between learners and the Chatbot service provider. Reference: Fuzheng ZHAO, Zibusiso MAFAITI, Yongjin LI, and Maiga CHANG. (2026). The research project aims to deploy MASA on Android platform with Android Studio and also to make improvements on ChatbotLLM service provider as well as related Generative AI and Natural Language Processing based research (see https://maiga.athabascau.ca/#outcome).

Research area, student roles & skills

Research area: One of the three research directions Dr. Chang is working on continuously is Data Analytics Service. In this research direction, his research focus on the design of algorithms and methodologies for learning activity generation, behaviour pattern extraction, and recommendation services. Dr. Chang is also appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang) working on Natural Language Processing related research that can automatically summarize the text the machine read and learnt for users' questions (see http://maiga.athabascau.ca/#askcovidq and https://ask4summary.vipresearch.ca), IEEE Northern Canada Section’s Capstone Project Award winnder project, ChatbotLLM (https://chatbot.vipresearch.ca/) that train chatbots on the materials uploaded.

Student roles:
The detailed function requirements will be given later when the students come to Canada. The student needs to use Android Studio with either Kotlin or Java to develop the mobile app as well as follow Google Play's policy and requirements to prepare necessary materials for publishing and updating the app. The student needs to fix the issues brought up by Google Play while publishing the app for VIP Research Group. Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps (including the mobile app) to access. Any functions that requires to use Python scripts beed to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

The student will be able to learn practical modern Python-enabled/supported web system architecture and implementation based on
• front-end browser-side JavaScript, CSS and HTML;
• making AJAX calls to access specific PHP endpoints (i.e., PHP pages);
• bridging by PHP and MariaDB;
• backend service in Python (with or without MariaDB) executed by and respond to PHP calls; and,
• data packaging with JSON for the communications among three layers.

A pilot study is needed to verify the effectiveness of the proposed method and the usability of the app; therefore, the student can know how to design a practical experiment and what statistical research methods can be applied to such research. Skills of designing and evaluating a questionnaire's validity and reliability can be learnt either.

Skills required:
For doing this project, the student is better to have the following skills or knowledge:
• [Required] Having experience in developing app with Android in either Kotlin or Java.
• [Required] Having experience in developing web app with PHP, JSON, HTML, CSS, and AJAX calls.
• Familiar with Python.
• Familiar with Git commands.
• Familiar with MariaDB and SQL commands.
• [Asset] Capable of adopting Responsive Web Design in web UI design.
• [Asset] Having experience in Network Socket and/or WebSocket programming.
• [Asset] Having experience in using Natural Language Processing library (NLTK) and LLMs in Python.

18. Mobile Accessible Weather Information Service and Dashboard based on Open Street Map

The research project involves two parts: (1) The production server needs to (a) have Open Street Map Tile Server for selected countries and regions; (b) allow administrators to upload Weather Information datasets; (c) preprocess the uploaded dataset based on 5-step KDD process; (d) train machine learning models based on the processed dataset or retrieve rules from the processed dataset; (e) allow administrators to test trained models or extracted rules by giving coordinates and date/time stamp; and (f) offer endpont for accessing weather information. (2) The Mobile Accessible browser app allows users to (a) see Open Street Map with weather overlays (could refer https://zoom.earth/); (b) zoom-in/out and click a place to see the relevant weather information; (c) click play button to see the animation of the weather changes. This research's outcome will be a weather information service provider for other research projects. Moreover, the weather and wildfire prediction and simulation functions could be built on the production server and the browser app later. Last but not the least, the mobile accessible browser app can be further extended to become a mobile device Android App with the support of Text-Generation, Speech-to-Text, and Text-to-Speech (TTS) GGUF models. Users can interact with the Android App via talking and listening.

Research area, student roles & skills

Research area: Three research directions Dr. Chang is working on are 1. Game-based Learning and Assessment (i.e., Stealth Assessment): have designed and developed various educational games on different platforms include desktop computers, smartphones, and televisions to assist users learning history and culture, programming language, financial concepts, and help Autism children on activities of daily living as well as elders with disabilities. 2. Data Analytics Service: focus on the design of algorithms and methodologies for learning activity generation, behaviour pattern extraction, and recommendation services. 3. Natural Language Processing: have been appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang).

Student roles:
The detailed function requirements will be given later when the students come to Canada. Based on interview (if there is any) discussions, the student may work on related Generative AI and Natural Language Processing based research (see https://maiga.athabascau.ca/#outcome) or other research project tasks that the student's skill set is fitting better.

The student needs to use browser-side JavaScript, CSS, HTML, and server-side PHP (the use of server-side JavaScript like NodeJS and ExpressJS is **not** allowed). Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps to access. Any functions that requires to use Python scripts beed to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

The student will be able to learn practical modern Python-enabled/supported web system architecture and implementation based on
• front-end browser-side JavaScript, CSS and HTML;
• making AJAX calls to access specific PHP endpoints (i.e., PHP pages);
• bridging by PHP and MariaDB;
• backend service in Python (with or without MariaDB) executed by and respond to PHP calls; and,
• data packaging with JSON for the communications among three layers.

A pilot study is needed to verify the effectiveness of the proposed method and the usability of the app; therefore, the student can know how to design a practical experiment and what statistical research methods can be applied to such research.

Skills required:
For doing this project, the student is better to have the following skills or knowledge:
• [Required] Experience on developing web app with PHP, JSON, HTML, CSS, and AJAX calls.
• [Required] Experience on developing independent application in Python.
• [Required] Experience on Machine Learning model training with Python.
• Familiar with Git, MariaDB and SQL commands.
• [Asset] Capable of adopting Responsive Web Design in web UI design.
• [Asset] Experience on executing Python script within PHP.
• [Asset] Experience on installing Open Street Map Tile Server on Ubuntu 24.04.
• [Asset] Experience and/or knowledge on Weather Information data.

19. Trading Card Game - Responsive Web Design and Android App

The project is enhancing the following existing research VIP Research Group (https://www.youtube.com/@vipresearchgroup) has: - The Trading Card Game v3.0 (you can watch the promotion video https://www.youtube.com/watch?v=MfYH5j_ge6U and the introduction video (https://www.youtube.com/watch?v=-vBV9_lHdy8 for 3.0 beta). After you watch the introduction video, you can register an account at https://tcg.game-server.ca/ and try to play against with yourself on another browser at https://play.tcg.game-server.ca/ Educational games have been proved that can be positively influence student's learning motivation and enhance student learning effectiveness. Researchers have also found that the use of in-game items as rewards for students according to their performance on learning activities can help to improve their academic achievements no matter they are boys or girls, are high-achievement or low-achievement students, and have high or low intention game players. Trinity CE Primary School at UK was inspired by my existing game-based educational reward system, Trading Card Game, and developed their own non-digital game "Character Clash!" to encourage children reading more. This project aims to analyze, design and implement the following features for the existing game-based educational reward system (https://tcg.game-server.ca) to help improving the game’s retention rate: 1. Card Forge System – students can use the earned points through game-play and the given cards by teachers to improve and get a higher quality card so they are willing to do more learning activities as well to play more. This project will contribute to educational game and game-based learning system R&D. Relevant functions have to be developed in modularized PHP pages to give flexibility and reusability in future development and enhancement for working with other systems. 2. Responsive Web Design (RWD) based webpage improvements. 3. Develop a WebView based mobile game client with Android Studio.

Research area, student roles & skills

Research area: Three research directions Dr. Chang is working on are 1. Game-based Learning and Assessment (i.e., Stealth Assessment): have designed and developed various educational games on different platforms include desktop computers, smartphones, and televisions to assist users learning history and culture, programming language, financial concepts, and help Autism children on activities of daily living as well as elders with disabilities. 2. Data Analytics Service: focus on the design of algorithms and methodologies for learning activity generation, behaviour pattern extraction, and recommendation services. 3. Natural Language Processing: have been appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang).

Student roles:
The detailed function requirements will be given later when the students come to Canada. Based on interview (if there is any) discussions, the student may work on related Generative AI and Natural Language Processing based research (see https://maiga.athabascau.ca/#outcome).

The student needs to use Android Studio with either Kotlin or Java to develop the mobile app as well as follow Google Play's policy and requirements to prepare necessary materials for publishing and updating the app. The student needs to fix the issues brought up by Google Play while publishing the app for VIP Research Group. Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps (including the mobile app) to access. Any functions that requires to use Python scripts beed to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

The student will be able to learn practical modern Python-enabled/supported web system architecture and implementation based on
• front-end browser-side JavaScript, CSS and HTML;
• making AJAX calls to access specific PHP endpoints (i.e., PHP pages);
• bridging by PHP and MariaDB;
• backend service in Python (with or without MariaDB) executed by and respond to PHP calls; and,
• data packaging with JSON for the communications among three layers.

A pilot study is needed to verify the effectiveness of the proposed method and the usability of the app; therefore, the student can know how to design a practical experiment and what statistical research methods can be applied to such research. Skills of designing and evaluating a questionnaire's validity and reliability can be learnt either.

Skills required:
For doing this project, the student is better to have the following skills or knowledge:
• [Required] Having experience in developing app with Android in either Kotlin or Java.
• [Required] Having experience in developing web app with PHP, JSON, HTML, CSS, and AJAX calls.
• [Required] Having experience in Network Socket and/or WebSocket application development.
• [Required] Capable of adopting Responsive Web Design in web UI design.
• Familiar with Git commands.
• Familiar with MariaDB and SQL commands.
• [Asset] Having experience in developing multi-client socket server in Java.
• [Asset] Familiar with Python.

20. VIP Translator Android App

The "VIP" stands for VIP Research Group (https://www.youtube.com/@vipresearchgroup). The VIP Translator project involves two parts: (1) the production server that offer endpoints for client apps (including the VIP Translator Android App) to send translation request and receive in a secure and privacy protection mechanics as well as allows administrators to test and verify the translation results on different and new Large Language Models served by Google, DeepSeek, OpenAI, etc. and open source instruction-tuned Large Language Models manually uploaded by administrators or automatic downloaded by the code; (2) the VIP Translator Android App allows users to switch BETWEEN the use of online translation production server via endpoints AND offline use of chosen model downloaded directly. For the latter option, the app needs to retrieve available models according to the output of an endpoint on the production server. The app can take NoteTakers' secure communication workflow and design as reference. The research project is based on existing research, see https://maiga.athabascau.ca/#notetakers: - The Generative AI (GenAI) supported Note Takers App (https://notetakers.vipresearch.ca), is an innovative tool designed to help users capture and organize notes from spoken conversations. In the case the users would like to use server-side speech recongition feature due to the built-in speech-to-text model is not always performing so well like a larger and mature open source model, the users also no need to worry about leaking any audio recording in the transit. It is because the Note Takers project incorporates Public Key Infrasture (PKI), Secure Sockets Layer (SSL) and salted secured hash algorithm to design and implement a privacy-protected and secure transmission mechanics and workflow.

Research area, student roles & skills

Research area: Dr. Chang are working on continuously on three directions: AI and Chatbot, Data Analytics Service, Mobile App and Game Development. Dr. Chang was appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang) working on Natural Language Processing related research that can automatically summarize the text the machine read and learnt for users' questions (see http://maiga.athabascau.ca/#askcovidq and https://ask4summary.vipresearch.ca), IEEE Northern Canada Section’s Capstone Project Award winnder project, ChatbotLLM (https://chatbot.vipresearch.ca/) that train chatbots on the materials uploaded.

Student roles:
The detailed function requirements will be given later when the students come to Canada. Based on interview (if there is any) discussions, the student may work on related Generative AI and Natural Language Processing based research (see https://maiga.athabascau.ca/#outcome).

The student needs to use Android Studio with either Kotlin or Java to develop the mobile app as well as follow Google Play's policy and requirements to prepare necessary materials for publishing and updating the app. The student needs to fix the issues brought up by Google Play while publishing the app for VIP Research Group. Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps (including the mobile app) to access. Any functions that requires to use Python scripts beed to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

The student will be able to learn practical modern Python-enabled/supported web system architecture and implementation based on
• front-end browser-side JavaScript, CSS and HTML;
• making AJAX calls to access specific PHP endpoints (i.e., PHP pages);
• bridging by PHP and MariaDB;
• backend service in Python (with or without MariaDB) executed by and respond to PHP calls; and,
• data packaging with JSON for the communications among three layers.

A pilot study is needed to verify the effectiveness of the proposed method and the usability of the app; therefore, the student can know how to design a practical experiment and what statistical research methods can be applied to such research. Skills of designing and evaluating a questionnaire's validity and reliability can be learnt either.

Skills required:
For doing this project, the student is better to have the following skills or knowledge:
• [Required] Having experience in developing app with Android in either Kotlin or Java.
• [Required] Having experience in developing web app with PHP, JSON, HTML, CSS, and AJAX calls.
• [Required] Having experience in developing independent application in Python.
• Familiar with Git commands.
• Familiar with MariaDB and SQL commands.
• [Asset] Capable of adopting Responsive Web Design in web UI design.
• [Asset] Having experience in using Natural Language Processing library (NLTK) and LLMs in Python.

21. Virtual Informal Personal (VIP) Chatbot Android App

The project is based on existing research, see https://maiga.athabascau.ca/#chatbotllm, VIP Research Group (https://www.youtube.com/@vipresearchgroup) has: - ChatbotLLM (https://chatbot.vipresearch.ca), is IEEE Northern Canada Section’s Capstone Project Award winnder project, is a system to automatically train chatbots on the materials uploaded by users. ChatbotLLM is not only capable of learning from material in traditional formats like PDF, PPT(X), DOC(X), and TXT, but also can reading XLS(X), image-scanned based PDF, images in PDF, DOC(X), and PPTX, and even MP3 with the help of Google’s Speech Recognition library. Reference: Supun De Silva and Maiga Chang. (2025). ChatbotLLM – Training Educational Chatbots on the Materials Uploaded by Teachers. In: 13th International Conference on Information and Education Technology (ICIET 2025), Fukuyama, Japan, April 18-20, 2025. DOI: 10.1109/ICIET66371.2025.11046253 (IEEE) Access: https://maiga.athabascau.ca/publication/Conference-2025-ICIET2025-ChatbotLLM.pdf - Mobile Accessible Seamless Agent (MASA) (https://masa.chatbot.vipresearch.ca), removes authentication barriers such as complex logins ensures instant access to learning materials on students’ devices via a familiar, intuitive interface. MASA’s UI/UX follows common mobile app patterns to minimize cognitive load and support natural, companion-like interaction between learners and the Chatbot service provider. Reference: Fuzheng ZHAO, Zibusiso MAFAITI, Yongjin LI, and Maiga CHANG. (2026). The research project aims to develop a mobile app with Android Studio and open source instruction-tuned Large Language Models. The mobile app could allow users to switch BETWEEN the use of online ChatbotLLM service provider while filtered the conversation (by setting by-pass to false) by subject content or not (by setting by-pass to true) AND offline use of chosen model downloaded directly. For the latter option, the app needs to retrieve available models according to the output of an endpoint on the production server. The app can take MASA's UI design as reference. The production server needs to have function and UI to allow administrator to test new models and auto update available model list.

Research area, student roles & skills

Research area: One of the three research directions Dr. Chang is working on continuously is Data Analytics Service. In this research direction, his research focus on the design of algorithms and methodologies for learning activity generation, behaviour pattern extraction, and recommendation services. Dr. Chang is also appointed as IEEE Computer Society Distinguished Visitor on AI and Chatbot (2023~2025, https://www.computer.org/profiles/maiga-chang) working on Natural Language Processing related research that can automatically summarize the text the machine read and learnt for users' questions (see http://maiga.athabascau.ca/#askcovidq and https://ask4summary.vipresearch.ca), IEEE Northern Canada Section’s Capstone Project Award winnder project, ChatbotLLM (https://chatbot.vipresearch.ca/) that train chatbots on the materials uploaded.

Student roles:
The detailed function requirements will be given later when the students come to Canada. Based on interview (if there is any) discussions, the student may work on related Generative AI and Natural Language Processing based research (see https://maiga.athabascau.ca/#outcome).

The student needs to use Android Studio with either Kotlin or Java to develop the mobile app as well as follow Google Play's policy and requirements to prepare necessary materials for publishing and updating the app. The student needs to fix the issues brought up by Google Play while publishing the app for VIP Research Group. Relevant functions supported from the production server have to be developed in modularized PHP to give flexibility and reusability in future development and enhancement for working with other systems and offered as endpoints for client apps (including the mobile app) to access. Any functions that requires to use Python scripts beed to be also implemented in modularized way and be able to execute solely with command line parameters for testing purpose as well as to be executed from a PHP page.

The student will be able to learn practical modern Python-enabled/supported web system architecture and implementation based on
• front-end browser-side JavaScript, CSS and HTML;
• making AJAX calls to access specific PHP endpoints (i.e., PHP pages);
• bridging by PHP and MariaDB;
• backend service in Python (with or without MariaDB) executed by and respond to PHP calls; and,
• data packaging with JSON for the communications among three layers.

A pilot study is needed to verify the effectiveness of the proposed method and the usability of the app; therefore, the student can know how to design a practical experiment and what statistical research methods can be applied to such research. Skills of designing and evaluating a questionnaire's validity and reliability can be learnt either.

Skills required:
For doing this project, the student is better to have the following skills or knowledge:
• [Required] Having experience in developing app with Android in either Kotlin or Java.
• [Required] Having experience in developing web app with PHP, JSON, HTML, CSS, and AJAX calls.
• Familiar with Python.
• Familiar with Git commands.
• Familiar with MariaDB and SQL commands.
• [Asset] Capable of adopting Responsive Web Design in web UI design.
• [Asset] Having experience in Network Socket and/or WebSocket programming.
• [Asset] Having experience in using Natural Language Processing library (NLTK) and LLMs in Python.