Showcasing undergraduate data science skill

Posted on March 04, 2019


QMIND EXECUTIVE: Managing Director of Operations Chaz Sutton, Founder and Managing Director Cooper Midroni, and Managing Director of Education Monica Cowper.

Queen's Machine Intelligence & Neuroevolution Design (QMIND) is scheduled to host the inaugural Canadian Undergraduate Conference on Artificial Intelligence (CUCAI) in Kingston, March 9.   

The event will feature talks, panel discussions, and workshops designed to help attendees explore and envision new, real-world applications for machine learning techniques. Perhaps most importantly, it’s a showcase for undergraduate students to network with each other and with industry representatives. Organizers are expecting as many as 200 participants.

“We have really begun to demonstrate to industry what undergraduate students and new graduates are capable of learning and applying in machine learning,” says QMIND Managing Director of Operations, Chaz Sutton. “QMIND members are developing real, working projects with good potential to create value for companies.”  

Most of the foundational mathematical and computational concepts for machine learning are decades old but, until recent years, technological limitations confined practical application to the realms of theory and high-level research laboratories. Now, the technology to cheaply and easily store and access huge data sets is available to virtually everyone. Powerful computer hardware with parallel processing capabilities is becoming cheap and plentiful. There is a rapidly growing range of accessible and user-friendly developer tools and platforms that allow enthusiast- and undergraduate-level developers to effectively use complex algorithms with just a few lines of code.  

All this means companies in all sectors are scrambling to hire employees with skills in machine learning. It’s not just a nice-to-have for employers; it’s a matter of competitive necessity. There are about 1,000 active job postings calling for applicants with data science skills across the country just this week.  

“Many people who are getting these jobs are trying to transfer laterally into data science from other established career fields,” says QMIND Managing Director and Founder, Cooper Midroni. “An applicant who has original, unique, and demonstrable project experience is likely to be considered highly in the applicant pool.” 

Midroni says that more than 90 students participate in 20 distinct QMIND projects. Nearly half the students work with data scientists associated with industry partners or as pro bono consultants for organizations in the Kingston community. Students in remaining projects explore their passions and interests while sharpening their skills. 

By way of examples, members of QMIND’s Automated Product Identification Team are working on a project that will detect, localize and track stock items in an on-campus grocery store. The system could eliminate the need to scan items on checkout. A scaled-up and refined system developed into a marketable product could mean huge operational savings for retailers.  

Members of QMIND’s Delirium Detection Team are working with physicians at Kingston General Hospital to develop and train an AI model to analyze various patient data to predict which patients may be at increased risk of delirium. Delirium is an often-rapid disturbance in mental ability often associated with a complex web of contributing factors. If doctors can better understand risk factors and identify patients most likely to suffer from delirium before they present symptoms, they can develop more proactive treatments, reduce mortality, and preserve quality of life.  

Members of QMIND’s Recycling Identification Team are developing a system to more accurately classify plastics in the recycling process. It’s technology that could reduce waste, further protect the environment, and reduce costs for municipal recycling programs.   

QMIND organizers are also developing a specialized 15-week course to cover some data science fundamentals entitled A Brief Introduction to Artificial Intelligence. It’s a further effort to strengthen the appeal of QMIND students to prospective employers.  

“We created a course to help students better understand the theory behind some of these algorithms," says Managing Director of Education, Monica Cowper. “With that information they can better see the possibilities, benefits, strengths, and weaknesses of the various tools and techniques; when to use learn machine learning and when not to.”  

Course topics include search trees, space graphs, state spaces, search algorithm properties, constraint satisfaction problems, game trees, complex decision making, reinforcement learning, probabilistic reasoning over time, Bayes networks, perceptrons, and kernels and clustering.  

The course is student-designed and executed, not offered by Queen’s University, and not part of any accredited engineering program, but that could change in the future. Demand is high from students and industry. Cowper says QMIND is in the preliminary stages, along with Queen’s Faculty of Engineering and Applied Science professors and administrators, of exploring how to develop an accredited undergraduate degree-level course for students.



Group 1

BIG INTEREST: QMIND has grown rapidly to more than 90 members since it inception in earlu 2017. This shot of new members from onboarding training in October, 2018.

Canadian Undergraduate Conference on Artificial Intelligence (CUCAI) 

Saturday, 9 March, 2019 

Four Points Sheraton Hotel 

285 King St, E. 



Find more conference information including a list of presenters here. 


QMIND Automated Product Identification team: 

Project Lead: Max Berkowitz 

Team Members: Logan Wilkinson, Justin Harrison, Tobias Carryer, Shreyansh Anand  


QMIND Delirium Detection team:

Project Lead: Stefano Roque 

Team Members: Deen Choudhury, Brendan Kolisnik, Jacob Beallor, and Jack Demeter 


QMIND Recycling Identification team: 

Project Lead: Marcus Uhthoff 

Team Members: Erik Beier, Joseph Grosso, Leon Zhou, Braeden Ng