Linkedin

Become a Member >

Uncategorized

How Artificial Intelligence is Reshaping Qualitative Research

AI tools offer new and exciting possibilities for analyzing complex social phenomena by interpreting language, images, and behavior—and they also raise important methodological questions.

On November 20, the Data Sciences Institute hosted a workshop exploring the ways that artificial intelligence is reshaping qualitative research. This was an exciting chance to connect at the intersection of AI, data science and the social sciences, an area that is ripe for further research attention. AI, and the data as well as data science that make it possible, are playing increasingly important roles in policy and decision-making, underscoring the need for robust engagement with how these technologies can be applied to social phenomena.

Co-organizers Professors Ethan Fosse (Associate Director, Data Sciences Institute, Department of Sociology, University of Toronto Scarborough) and Nicholas Spence (Departments of Sociology and Health & Society, University of Toronto Scarborough) wanted to create a space to foster collaboration and interdisciplinary innovation between the computational and social sciences.

Their aim is to build on the momentum of what Fosse dubbed “the Toronto Revolution,” or the era following Geoffrey Hinton’s arrival at the University of Toronto, which has cemented the university’s status as the birthplace of the current AI boom. In addition to the collaborative potential as new tools help qualitative researchers obtain new insights from unstructured data, the workshop also aimed to tackle the ethical and methodological questions that these tools raise.

“The Toronto Revolution did more than improve image recognition,” says Professor Fosse. “By rendering unstructured data computationally accessible, it has enabled the infrastructure to finally bridge (and dissolve) the historic divide between qualitative and quantitative research in the social sciences. The objective is augmentation, not automation — using computational tools to extend, rather than replace, traditional qualitative research methods.”

Professor Susan McCahan (Department of Mechanical and Industrial Engineering. Faculty of Applied Science and Engineering; and Associate Vice President & Vice Provost, Digital Strategies, University of Toronto) explored some of these methodological questions by presenting methodologies used in her research. Still, she said, the question remains what new AI methods need to be developed given the rapidly changing technologies.

“Even if you could find a list of AI capabilities in the literature, it would need to be updated every day, given how fast as things are changing.”

Professor Corey Abramson (Sociology, Rice University) delivered the keynote, “On Patterns, Edge Cases, and Scalable Interpretation: Pragmatic Uses of AI in Qualitative Research.” He argued that, from filing cabinets to qualitative data analysis tools, fraught technologies are nothing new. AI technologies, too, are both generative and dangerous—neither all good nor all bad. While the wide breadth of qualitative research will mean AI has different relevance for different researchers, he argued that it is essential for all to at least understand it. After all, this technology is shaping the social world that researchers seek to understand.

Dr. Qin Liu (Senior Research Associate, Institute for Studies in Transdisciplinary Engineering Education and Practice, Faculty of Applied Science & Engineering, University of Toronto) similarly emphasized the continued role of humans in understanding the social world. She argued for maintaining human engagement with qualitative data and with the results of AI coding. While such coding can augment researchers’ capacity for certain types of analysis, human intelligence is needed to bridge gaps between different ways of thinking about the world.

Dr. Jordan Joseph Wadden (Clinical Ethicist, Centre for Clinical Ethics, Unity Health Toronto; Assistant Professor (status), Department of Family and Community Medicine, Temerty Faculty of Medicine, University of Toronto) highlighted ethical factors that developers, researchers, and policymakers must contend with, emphasizing that such ethical considerations are an opportunity to reflect and improve research and create positive impacts by identifying and addressing problems early.

Professor Spence echoed this sense that careful consideration makes qualitative research incorporating AI better. This integration, he noted, “is being done in great, published research.” Spence continued: “Humans have not been removed from the process” but rather have an “additional set of tools to look at social world in new ways.”

This workshop is a step towards establishing U of T as a leader in AI-assisted qualitative research. The organizers look to continue its momentum with further workshops bringing together an even greater breadth of researchers.

“U of T possess a high density of world-class qualitative scholars in the social sciences and other fields, and we inherit the legacy of the Toronto Revolution in deep learning,” concluded Professor Fosse. “By merging these strengths, we can establish the University of Toronto as a leader in rigorous, transparent, and theoretically grounded AI-assisted qualitative research.”

 

Deploying AI Microcredential Sees Soaring Interest from Professionals

The Data Sciences Institute’s Deploying AI microcredential is a massive success in its first session. Organizations are seeking employees who can deploy AI responsibly and effectively—and professionals are responding with incredible demand for learning opportunities to build their skills.

Deploying AI is a hands-on opportunity to dive into the power of Large Language Models (LLMs). Learners develop the technical know-how and practical strategies needed to take AI from prototype to production.

Responding to input from learners and employers, DSI launched Deploying AI this fall and the enthusiasm from learners was palpable. The offering quickly sold out, reflecting soaring demand for the tools and confidence to deploy generative AI at scale.

The microcredential builds on the success of the DSI Data Science and Machine Learning Software Foundations Certificates offered with the financial support of Upskill Canada, powered by Palette Skills and the Government of Canada. For learners that have completed a Certificate, this microcredential is a next step in deepening their AI capabilities.

Deploying AI offers a unique value as it is short, targeted learning experience dedicated to the frameworks, tools, and applied skills that professionals need to navigate the ethical, operational and organizational challenges of AI integration.

Deploying AI incorporates case studies from industry. Recently, learners heard from Sepehr Sisakht, an industry leader who is applying AI in practice as CEO of Shyftbase Inc. Learners gained a firsthand look at how AI deployment challenges are approached in a real-world setting. Applied perspectives like this bridge the gap between technical training and practical implementation, as learners gain insight into the workflows, trade-offs, and decisions faced by professionals deploying AI at scale.

“We developed Deploying AI in response to industry demand. Employers need people who can deploy and scale AI solutions in the real-world, now,” says Prof. Rohan Alexander, Director, Technical Skills and Curriculum (Faculty of Information and Department of Statistics, Faculty of Arts & Science).

“We’ve heard from our learners that our microcredential is making a real difference, whether they are preparing for that next great job or are in a role where they can put their learnings into practice right way. The response has been phenomenal.”

That response is echoed in the feedback on Deploying AI from learners. Learners highlighted the immediate usefulness of the microcredential, noting that the microcredential “provided a great primer to the tools and concepts for building AI solutions” and that the theory was useful and the practical knowledge immediately applicable to my work.”

Learners also emphasized the value of hands-on projects, noting that the assignments allowed them to showcase their ability and build credibility as AI professionals.

Registration is open for the February session of the Deploying AI microcredential. Don’t miss your chance to build the skills employers need to deploy AI effectively.

Learn more and register

Helping prepare Canada’s financial services workforce with AI literacy

How U of T’s Data Sciences Institute is helping professionals and employers stay competitive

As digital literacy becomes crucial to navigating the financial sector, the rapid pace of change can feel overwhelming for many mid-career experts and their employers, alike. This is particularly true in areas such as data sciences, artificial intelligence (AI) and machine learning. 

Hiring professionals who are proactively adapting to the rise of data science and generative AI is one of the smartest ways businesses can stay competitive. 

The University of Toronto’s Data Sciences Institute (DSI), a hub for data science research, training and partnerships, is helping businesses do just that by growing a pool of job-ready professionals. The DSI has upskilling certificates and microcredentials targeting the growing demand for professionals in finance and other fields to become knowledgeable in AI and understand how available tools can enhance their work.  

Since launch, over 500 professionals have completed certificates, forming a job-ready talent pool aimed at bridging the gap between skilled candidates and employers. The DSI plays a vital role and is dedicated to creating new career paths for untapped talent to unleash their full potential. 

DSI learners have access to career transition support and exclusive employer networking events.

As a manager of anti-money laundering and compliance data at Scotiabank, Matias Velastegui is one such professional. In search of a way to boost his technical skills, he completed Machine Learning Software Foundations, a 16-week intensive certificate at the DSI.  

With the financial support of Upskill Canada, powered by Palette Skills and the Government of Canada, DSI certificates and microcredentials train mid-level professionals on digital literacy skills, including AI and machine learning, and are designed to meet the talent needs of high-growth sectors. 

“It provided me with valuable tools that I’m confident I’ll apply in future professional and academic projects,” Velastegui says.  

Certificates and microcredentials at the DSI are delivered live online, with weekly support through virtual office hours. Included are opportunities to learn from industry experts during case studies that provide participants with important insights into the professional world of data science and AI analytics.  In parallel to the technical training, learners have access to career transition support and exclusive employer networking events, which strengthens talent pipelines and helps keep Canadian businesses globally competitive. 

While the certificate offers a comprehensive data science and machine learning overview, the DSI also offers a shorter, three-week microcredential on Deploying AI that focuses on building AI applications to augment work tasks. Deploying AI provides professionals with technical and strategic skills to turn AI prototypes into practical, workplace-ready solutions. 

Since its launch, over 500 professionals have completed DSI certificates.

Velastegui says he decided to pursue the DSI certificate because it combined academic rigour with a practical focus. In his current role, he works extensively with SQL and Python languages. Through the DSI certificate, he learned new strategies for approaching complex queries, more efficient ways of handling data and best practices for data governance.  

“From ensuring data integrity to grasping the mathematical principles of neural networks, professionals must be actively engaged in these advancements and prepared to evolve alongside them,” Velastegui says.  

Now, he plans to hire other DSI upskilling participants to his own team. Velastegui is confident in the DSI’s focus on improving coding expertise and collaboration skills in virtual environments.  

The DSI offers employers looking for new talent the opportunity to share job opportunities with its pool of participants, all of whom have post-secondary degrees and now have enhanced skills in data science and machine learning.  

The certificates in Data Science and Machine Learning Software Foundations offer part-time coursework, so participants can continue working regular hours. Sakib Sadat, an intelligent automation program manager at Manulife, says the certificate is intensive but worth the time commitment.  

Sadat says there has been a strong industry push for professionals to boost their skills using AI and machine learning tools, both for company standards and for personal development. Participating in the DSI’s certificate was a way for him to improve his productivity and the quality of his work.  

Through the certificate, Sadat gained the ability to critically assess and identify which AI solutions presented by clients add real value in his current role. 

“This certificate was one way to get the accumulated technical expertise to make those assessments on whether or not AI is going to actually be productive,” Sadat says. 

Beyond the technical skills, Sadat and Velastegui agree that adding a University of Toronto AI-related certification to your resume is a good way to stand out in a pool of professionals, either when looking to move up internally or make a career switch.  

“From a career progression standpoint, AI is a personal brand and selling feature in the industry, whether you’re in IT or not,” Sadat adds.  

For professionals looking to learn more about the DSI’s upskilling offerings and employers accessing the DSI professional talent pool, visit certificates.datasciences.utoronto.ca.  

This story was created by Content Works, Postmedia’s commercial content division, on behalf of the Data Sciences Institute. 

AI in Genomics: Building a Collaborative Future for Health Innovation

Genomics and data science researchers from across the University of Toronto (U of T) and affiliated research institutes recently gathered to explore a shared frontier: how artificial intelligence (AI) can accelerate discoveries in genomics and improve patient outcomes.

AI in Genomics – a community-building day of presentations and discussions designed to explore how AI can help unlock new insights across genomics and multi-omics research – was a collaborative event co-sponsored by the SickKids Research Institute, the McLaughlin Centre and the Data Sciences Institute.

“Our goal is to highlight the exciting work already happening across the U of T ecosystem and to create space for meaningful connections,” said Lisa Strug, Director, Data Sciences Institute and Professor in the Departments of Statistical Sciences and Computer Science (Faculty of Arts & Science) and the Division of Biostatistics (Dalla Lana School of Public Health) at the University of Toronto.

“We’re looking to seed new collaborations and larger initiatives that will push this field forward and we are pleased to work with SickKids and McLaughlin on the event.”

“Detecting cell-cell communication from transcriptomic data is by nature extremely difficult, as the actual binding occurs at the protein level. As such, we required the use of AI-powered models to learn patterns too complex for other traditional models. To meet this challenge, we developed a specialized model trained in an unsupervised way and, instead of using the direct output of the model, we opened up the machinery of the model to detect cellular communication in a new way,” said Gregory Schwartz, one of the presenters and Canada Research Chair in Bioinformatics and Computational Biology and Scientist, Princess Margaret Cancer Centre, University Health Network Assistant Professor, Department of Medical Biophysics, University of Toronto.

“We have all been using AI and machine learning for years but everyone in their own way. In some cases, we are world leaders,” explained Stephen W. Scherer (Chief of Research, Northbridge Chair in Paediatric Research, Senior Scientist, Genetics & Genome Biology program, The Hospital for Sick Children (SickKids); Director, McLaughlin Centre, University of Toronto).

“At the Future of Homo Sapiens event we hosted at SickKids last fall, there was a memorable moment when Craig Venter and Geoffrey Hinton clashed over the potential impact, and risks, of AI. That sparked the idea for AI in Genomics as an extended ‘lab meeting’… which quickly evolved into something much bigger.”

Aiming to spotlight emerging research, spark interdisciplinary partnerships, and shape a growing community dedicated to the responsible and impactful use of AI in genomic science, AI in Genomics served as a platform for faculty, trainees, students, and research staff to share their work, learn from one another, and identify key opportunities where AI can address pressing challenges in genomics.

AI in Genomics encouraged participants to map out areas within genomics – such as disease risk prediction, gene expression analysis, or precision medicine – that could benefit most from advanced computational tools like machine learning and deep learning. The research panel explored impacts of AI in genomics from getting AI tools into the hands of clinicians and uses for optimizing population health. Researchers highlighted the need for a continuous cycle of discovery and implementation, and the need to figure out where is the right place for structured and unstructured data that can be used for research or clinical care, as well as the importance of reproducibility for research in AI in genomics.

“The DSI brings communities together to help advance fields,” Strug added. “This was supposed to be a small intimate event to understand what’s happening on campus but the demand reflected that this is already a major area of interest and opportunity. We hope to better understand what is happening, how we can fill training gaps and how we can support the community to advance this area and realize the limitless opportunities.”

Data Sciences Institute Celebrates SUDS Cohort of 2025 with Showcase

The Data Sciences Institute’s (DSI) Summer Undergraduate Data Science (SUDS) Opportunities Program celebrated the achievements of its 2025 cohort with the annual SUDS Showcase – an exciting full day of research project presentations and poster sessions by 60 undergraduate students.

Designed as a marquee event to close the SUDS year of study, the Showcase provides a forum for SUDS Scholars and Supervisors to share their data science research.   

Javier Mencia Ledo, SUDS 2025 Scholar, whose research Risk factors and Early Prediction of Labour Force Dropout in SLE Patients: Integrating Longitudinal Deep Learning through an LSTM RNN with Random Forests, focused on a neural network that detects early warning signs of disability in lupus patients, allowing timely support for interventions. Supervised by Professor Behdin Nowrouzi-Kia (Department of Occupational Therapy and Occupational Science, Temerty Faculty of Medicine), Javier worked in the Rehabilitation Sciences Through Occupational Research & Engagement (ReSTORE) Lab.

“Being part of SUDS has been such an invaluable experience,” said Ledo.

“I got to hear the stories and learn from incredibly talented people working in both industry and academia, and contribute to many impactful projects at the ReSTORE Lab. It confirmed that I want to pursue this career path in grad school.”

“The SUDS Showcase is a highlight, creating an opportunity for scholars, supervisors and the broader DSI community to view and discuss the various data science methods, including AI, applied across a broad range of areas,” said Professor Laura Rosella, DSI Associate Director of Education and Training.

“Under the supervision of U of T and affiliated external partner researchers, students applied data science methods and tools to research on locating genetic ancestors with ancient DNA, integrating predictive analytics into an equity dashboard and finding substructures within the Milky Way with geometric deep learning.

“Elise Corbin and Al Ali Abdulmohseen collaborated in the presentation, Piccard: An Open-Source Tool to Analyze Longitudinal Data without Geographic Harmonization, detailing the development of a Python package that applies graph networks to census data visualization and analysis. Their research was supervised by Professor Fernando Calderón Figueroa (Department of Human Geography, University of Toronto Scarborough).

“SUDS was like a dream job for me,” said Corbin. “I really enjoyed a flexible schedule, and I felt like I was doing important work that could really improve people’s lives down the line.”

“My collaborator, supervisor, and I hope to publish the results of our work as well, which is an added bonus. I recommend SUDS as the perfect opportunity to gain research experience, experience life in the data science workforce, and possibly even get published!”

(L-R) Matthew Tamura, Shan (Angelina) Zhai, Professor Shion Guha (Faculty of Information, University of Toronto) worked on the Children’s Aid Society MITACS project

SUDS provides a rich summer training experience for students from a wide variety of academic backgrounds to be exposed to and apply data science techniques in their work.

Two SUDS Scholars from the University of Toronto had the opportunity to intern at Children’s Aid Society of Toronto, thanks to Mitacs funding. This collaboration is part of the larger DSI initiative for Data-driven Decisions & Discovery: Innovation for Transformative Impact. Through these strategic partnerships, DSI connects organizations with skilled undergraduate talent to advance high-impact, data-driven projects. With Mitacs support, partners can accelerate innovation by engaging top U of T students over the summer.

“The MITACS Accelerate program has been instrumental in bridging academic research with real-world impact,” said Prof. Shion Guha (Faculty of Information and Department of Computer Science, U of T).

“For example, through our partnership with the Children’s Aid Society of Toronto, two outstanding undergraduate SUDS Scholars are contributing to data-driven solutions in the child welfare sector, gaining invaluable experience while shaping socially responsible technology.”

The 27 students from the King Abdullah University of Science and Technology (KAUST) Academy, recipients of prestigious awards from KAUST, were selected through a highly competitive process to participate in SUDS. This marks a near doubling from last year’s SUDS KAUST cohort, reflecting growing interest and momentum. KAUST specifically sought out the University of Toronto for this collaboration due to its world-renowned ranking in data science.

“The SUDS Scholars were excellent, and it was great to see them present their research, building on the data science skills they have learned this summer,” said DSI supervisor Zahra Shakeri, (Dalla Lana School of Public Health, University of Toronto).

“They worked closely with the clinician in the team and other team members to explore a timely data science problem, providing valuable insights and framing directions for future investigation.”

Along with their research projects, SUDS Scholars partake of the SUDS Cohort programming for networking, academic and professional development. This includes the Data Science@Work Series, where representatives from the private sector and government organizations share data science applications in the workplace. The scholars began in May with the DSI Data Science Bootcamp, gaining proficiency in data science skills including Unix Shell, R, Python, and machine learning.

A highlight of the 2025 Showcase was keynote speaker, Prof. Rachel Harding (Department of Pharmacology and Toxicology, Temerty Faculty of Medicine, University of Toronto; Principal Investigator, Structural Genomics Consortium), who spoke on the topic of Protein–Ligand Data at Scale: Foundations for Machine Learning in Drug Discovery. Does that work? 

“The SUDS program offers a rare and powerful blend of technical training, critical thinking, and applied experience,” said Guha.

“As a faculty mentor, it’s been deeply rewarding to witness students grow into thoughtful, industry-ready researchers committed to ethical data science.

Distinction in the poster category was given to scholars Amjad Albawardi, Tabris Cao, Abdulaziz Alkharjy, Anas Alshehri and Mehtab Cheema, while Noor Khan, Matthew Tamara and Shan (Angelina) Zhai were recognized for their standout presentations.

Photos: Cormac Rea