Data Sciences Institute

by Sara Elhawash

Hotter days in Toronto mean more people flocking to parks for relief, but just how are these green spaces being utilized during extreme heat? University of Toronto researchers, who were awarded the Data Access Grant by the Data Sciences Institute, are analyzing patterns of human activity, park usage and air temperatures to shed light on the impact of extreme heat and climate patterns on the health and well-being of Toronto residents. 

by Sara Elhawash

The Data Sciences Institute (DSI) is excited to announce a new partnership with the Ontario Institute for Cancer Research (OICR), a collaborative research institute that conducts and enables high-impact translational cancer research.  

OICR conducts cross-disciplinary cancer research in areas such as genomics, immuno-oncology, informatics, computational biology, genome informatics, implementation science, drug discovery, and molecular pathology while facilitating global research collaboration, securely sharing data, and providing powerful, world-class tools and resources to the research community. 

“Our collaborative approach, both locally and globally, ensures that Ontario remains at the forefront of cancer research and care. With a shared commitment to maximizing the health and economic benefit of our research for the people of Ontario, this partnership with DSI holds tremendous potential to drive breakthroughs in cancer research that can bring real benefits to those affected by cancer,“ said Dr. Laszlo Radvanyi, President and Scientific Director, OICR. 

DSI collaborates with organizations eager to support world-class researchers, educators, and trainees advancing data sciences. We facilitate inclusive research connections, supporting foundational research in data science, as well as supporting the training of a diverse group of highly qualified personnel for their success in interdisciplinary environments.  

As one of the DSI external funding partners, OICR researchers with an appointment at the University of Toronto can apply for research grants, supports and training and lead initiatives at the DSI.  

“We are very excited to have the Ontario Institute for Cancer Research join our growing DSI community. Our goal is to create a hub to elevate data science research, training, and partnerships. By connecting and supporting data science researchers, the DSI advances research and nurtures the next generation of data- and computationally focused researchers,“ says Lisa Strug, Director, Data Sciences Institute. 

by Sara Elhawash

Hate speech and misinformation on social media can have a devastating impact, particularly on marginalized communities. But what if we use AI to combat such harmful content? That’s the goal of a team of University of Toronto researchers who were awarded a Catalyst Grant by the Data Sciences Institute to develop an AI system to address the marginalization of communities in data-centric systems – including social media platforms like Twitter. 

The collaborative research team, which consists of Professors Syed Ishtiaque Ahmed, Department of Computer Science (Faculty of Arts & Science), Shohini Bhattasali, Department of Language Studies (University of Toronto Scarborough) and Shion Guha (Faculty of Information), intends to make content moderation more inclusive by involving the communities affected by harmful or hateful content on social media. The project collaborates with two Canadian non-profit organizations: the Chinese Canadian National Council for Social Justice (CCNC-SJ) and the Islam Unravelled Anti-Racism Initiative. 

Professor Ahmed shares that historically marginalized groups are most affected by content moderation failings as they have lower representation among human moderators and their data is less available for algorithms. He says, “While most social media platforms have taken measures to moderate and identify harmful content and limit its spread, human moderators and AI algorithms often fail to identify it correctly and take proper actions.” 

The team plans to design, develop, deploy, and evaluate the proposed system to address potential Islamophobic and Sinophobic posts on Twitter. The AI system aims to democratize content moderation by including diverse voices in two primary ways. First, by allowing users to contest a decision, the moderation process becomes more transparent and trustworthy for users who are victims of online harms. Second, by taking user input and retraining machine learning (ML) models, the system ensures that users’ contesting positions reflect on the prescreening ML system. 

Ahmed explains, “Annotating data becomes challenging when the annotators are divided in their opinions. Resolving this issue democratically requires involving different communities, which is currently not common in data science practices. This project addresses the issue by designing, developing, and evaluating a pluralistic framework of justification and contestation in data science while working with two historically marginalized communities in Toronto.” 

The AI system will integrate the wisdom, knowledge, and experiences of community members into the process of reducing hateful content directed toward their communities. The team is using a participatory data curation methodology. They learn about the characterization of different kinds of harmful content affecting a community and include members of the corresponding community in the data labeling process to ensure data quality. 

“We are grateful to DSI for their generous support for this project. The DSI community has also helped us connect with people conducting similar research and learn from them. Thanks to the wonderful DSI community, whose mission includes  innovating and adopting various data-centric approaches to social justice,” says Ahmed. 

The research project is a promising initiative to address the issue of harmful content on social media and is expected to have far-reaching impacts beyond the two communities it is currently focusing on. 

by Adina Bresge

When Madeleine Bonsma-Fisher bikes through Toronto, she sees where her research meets the road.

Each street she pedals down presents as a series of data points: She’ll count 15 people weaving past one another on the sidewalk, while three cars cruise down a road that takes up 80 per cent of the space.

A cycling activist, Bonsma-Fisher is studying traffic patterns as part of her postdoctoral research at the Data Sciences Institute, an institutional strategic initiative that is a tri-campus hub for number crunchers across disciplines. Before that, she modelled evolutionary interactions between microbes.

The common thread? Data and data analysis.

“I don’t want to say that data science is the answer to everything, but I am finding that there is so much you can do,” Bonsma-Fisher says. “It gave me a lot of freedom to really just do whatever I wanted.”

Her current research focuses on what might seem like a simple question: At any point in Toronto, can you cycle to essential destinations – grocery stores, health care and schools – within 30 minutes, using only bike lanes and traffic-calmed roads?

The answer, she says, is far from straightforward. It requires sophisticated data analysis to make a map of the entire city and rate each road according to traffic stress, which accounts for factors such as traffic volume, speed limits and physical separation.

The next step, Bonsma-Fisher says, is to pinpoint places where infrastructure could improve access to cycling as a comfortable and convenient mode of transportation, such as dedicated bike lanes and physical separation from car traffic.

As she searches for active transportation solutions, Bonsma-Fisher is working with two advisers at the Data Sciences Institute: Shoshanna Saxe, an associate professor in the department of civil and mineral engineering, and Timothy Chan, a professor of mechanical and industrial engineering – both in the Faculty of Applied Science & Engineering.

“What’s cool about the Data Sciences Institute is that the vision is to bring people together with different experience and allow people to make that jump to a different field.”

The winding road of Bonsma-Fisher’s research career – and the data focus that underpins it – began when she arrived at U of T’s School of Graduate Studies in 2014 with a physics degree and an interest in using the field’s principles to solve biological problems.

Her supervisor, Sidhartha Goyal, an associate professor in the department of physics in the Faculty of Arts & Science, suggested she look into CRISPR – a term she hadn’t heard before, but one that would become the subject of both her master’s and doctoral studies.

You may have heard of CRISPR in the context of genome editing, but the technology is derived from a bacterial defence mechanism that is analogous to adaptive immunity in humans. Many bacteria have an immune system called CRISPR that allows them to store memories of viruses in their own DNA – like a genetic gallery of viral “mug shots,” Bonsma-Fisher explains.

As part of her PhD research, Bonsma-Fisher built a simple mathematical model to explore how computer-simulated interactions between populations of bacteria and viruses shape CRISPR immune memories.

The paper, published in the journal eLife earlier this year, provides fresh insight into the evolutionary “arms race” between viruses and bacteria – with viruses mutating to evade immune recognition, while CRISPR builds bacteria’s DNA database of previous attackers. The simplicity of the model helped narrow down the most prominent processes in a complicated system, Bonsma-Fisher says.

Down the road, Bonsma-Fisher says the model could contribute to our understanding of immunity in more complex organisms, including humans.

“Some of the conclusions we think are going to apply to any type of immune system-virus interaction.”

While she was chipping away at her microbial models, Bonsma-Fisher made another discovery: data analysis skills were in short supply – and high demand – among her fellow graduate students. So, she co-founded the U of T Coders group to give researchers across all disciplines a chance to learn the basics of programming and teach each other new techniques through hands-on, member-led tutorials.

“A lot of people would try to learn by themselves,” she says, “and there would be a lot of struggle and tears. U of T coders was a place for people to support each other through all of that.”

by Sara Elhawash

This summer, a group of 37 undergraduate students from across Canada will participate in the Data Sciences Institute (DSI) Summer Undergraduate Data Science (SUDS) research opportunity. SUDS offers an enriching summer experience to apply data science methods and tools in various application areas, including humanities, life science, engineering, public health and more. SUDS Scholars are supervised by DSI member researchers across U of T and external funding partners. In addition to their research projects, SUDS Scholars are provided with a full set of data science skills, networking, research and professional development opportunities. 

The SUDS Scholars kicked off their summer programming on May 1 by participating in their welcome and orientation where they met their cohort and shared their research interests. This week, they will also be gaining new skills and knowledge via the Data Science Bootcamp for introductions to Unix Shell, R, Python, and Machine Learning.   

SUDS Scholar Yuanhan Peng, expressed enthusiasm about the program, noting she is “currently strongly interested in education data science and I want to work as a data scientist in the education technology industry. I think that being a SUDS Scholar will provide me with the opportunity to participate in related research and gain more experience.”  

SUDS Scholar, Akil Huang, will be working with Professor Spike Lee, Rotman School of Management on the project, Automated text analysis of liberal and conservative news, which explores the nature of linguistic differences among news outlets. Huang is looking forward to the summer: “As a finance and economics specialist coming from an interdisciplinary background, SUDS will give me the tools, learning resources, and mentorship to delve deeper into my area of focus within academia. I’m really looking forward to the bootcamp in the beginning of the program.” 

Over the next twelve weeks, the Scholars will engage on their data science research projects and attend the DSI cohort programming. The programming includes the Data Science@Work Series where members from the private sector and government organizations discuss data science in the workplace.  

Professor Kuan Liu, Dalla Lana School of Public Health and SUDS supervisor shares, “The DSI SUDS program provides unique hands-on research and learning experiences in data science that help students gain exposure to cutting-edge research topics spanning a wide range of disciplines and develop critical scientific communication, computation, and problem-solving skills. It was a great pleasure working with the SUDS scholar this past summer and I am excited to take part in the program again this year.” 

The program will conclude in August with the DSI Research Day, where the SUDS Scholar cohort will showcase their research.

Professor Laura Rosella, DSI Associate Director of Education and Training, shares that “As we welcome the new cohort of students, we are excited to see their passion and curiosity. Besides engaging in research projects, our SUDS Scholars benefit from acquiring data science expertise and professional growth opportunities. We are enthusiastic about the prospect of inspiring these students and, hopefully, launching their careers in data science. They are an outstanding group indeed.”  

The SUDS research opportunity is an excellent way for students to get involved in high-quality and enriching data science learning and to experience the application of data science methods and tools in various fields. It provides an opportunity for students to build their careers in data science.