Projets novateurs réalisés

Evaluating Clustering Techniques using a range of Semantic Similarity Evaluation Metrics

This research project will allow us to assess how similar two news articles are to each other. This may be easy for a human to do, but for machines this is actually quite hard. OpenReg will make use of this assessment to determine if a news article is important for their users. The goal is to be able to identify to what degree 2 articles need to be similar to each other for the user to be interested, yet different enough for them to not read duplicate information.

Sing pray heal: Exploring the Health Benefits of Modern Kirtan as Sacred Community Music Therapy

This research project is an exploration of the salient health benefits of participating in chant-based music experiences, ultimately to determine whether or not such music making can have perceived therapeutic impact. Merging the fields of music therapy and medical ethnomusicology, I will examine kirtan chanting as ‘sacred community music therapy’ and utilize a five-dimensions model of health, grouping perceived benefit into physical, emotional, mental, social and/or spiritual domains.

Kirtan is a call-and-response style of prayerful singing that originated in East India and now presents in modern forms led by non-Indians, yogis and musicians. Shanti Mission, a group based in Cooranbong, Australia, uses modern kirtan as a primary component of their healing work. Through ethnographic fieldwork, I will engage in participant observation of Shanti Mission events, performances, workshops and classes and will conduct interviews with both musicians and participants. For comparison and contrast, I will also observe more ‘traditional’ kirtan in contexts such Hindu and Sikh temples in Edmonton. “TO BE CONT’D”

Calcul de la rigidité superfluide dans le régime de coexistence de phase dans les cuprates à température finie

Qu’est-ce que la supraconductivité non-conventionnelle? C’est un phénomène quantique se produisant à haute température dans certains matériaux et associé à la disparition de toute résistance électrique et par l’expulsion des champs magnétiques, une fois ces derniers refroidis sous une température critique Tc et en présence d’un champ magnétique inférieur à un champ magnétique critique Hc. Fondamentalement, la supraconductivité est caractérisée par la formation de paires de Cooper (paires d’électrons), en raison d’une force d’attraction effective entre les électrons. Je m’intéresse aux supraconducteurs à base d’oxyde de cuivre (souvent simplement appelés cuprates), découverts à partir de 1986, dont les Tc sont de plusieurs dizaines de degrés (le record étant de 150K). Dans cette classe de matériaux, de façon complètement contre-intuitive, ce serait les très fortes interactions coulombiennes (répulsives) entre les électrons qui seraient à l’origine de la supraconductivité. TO BE CONT’D

Kappa architecture serving layer suitability

Global service providers in highly regulated financial sectors must accommodate an ever-changing, sometimes competing, landscape of regulatory and business concerns. This project seeks to define a technology infrastructure design that supports current and anticipated data privacy and data residency concerns, making it possible to keep data within borders while still facilitating collaboration across those borders. Consumers are increasingly aware of the collection of their private data, but are often unaware of cross-border movement of their data. Businesses are cognizant of the limited (or untested) reach of legal and regulatory recourse when their data (and their customers’ data) unnecessarily crosses borders. This project seeks to add technology frameworks to aid in addressing those customer and business concerns.

De la vulgarisation scientifique à l’expertise des organisations à but non lucratif : les défis et enjeux du communicateur professionnel dans la production de contenu sensible grand public.

La vulgarisation scientifique suppose de préparer du contenu dit « expert » en vue d’être diffuser et lu par des publics non experts. Ce passage vers un contenu profane demande de comprendre l’information d’origine, mais également des différents publics auxquels il s’adressera : il faut alors communiquer un message pertinent pour ce public au regard de l’expérience du monde de ce destinataire; de ses jugements culturels et sociaux et du sentiment qu’il éprouve face au message et à son destinateur. Cette tâche est parfois réalisée par un journaliste, mais souvent par un professionnel de la communication qui œuvre dans une organisation. Aussi, en collaboration avec la Fondation Jasmin Roy, ce projet vise à comprendre les enjeux de la vulgarisation scientifique dans la production de contenu sensible, ici l’intimidation, la violence et la discrimination faites aux enfants en milieu scolaire au primaire et au secondaire. Ces enjeux peuvent être d’ordre organisationnel, techniques, technologiques ou rédactionnels.

Classification of radiological observation through image-sentence association

Imagia is an AI-driven personalized healthcare company, enabling collaborative development of predictive biomarkers. Its Evidens platform unites deep learning expertise and clinical insights on federated patient data from partnered hospitals & AI research institutions. Imagia delivers impactful solutions to healthcare providers, pharmaceutical companies and medical device manufacturers. Imagia’s mission is to leverage advances in artificial intelligence to accelerate accessible personalized medicine with an initial focus on personalized oncology.
This project aims at developing a software prototype which can automatically, or semi-automatically, associate available annotations to their context in radiological reports.

Storytelling the History of Cantonese Migration through Virtual Heritage

Since the 19th century, Chinese men and women have migrated to small towns and cities throughout places such as Southeast Asia, Australia, New Zealand and North America in search of better opportunities, bringing with them many traditional cultural practices, values and customs that evolved over time and space. How did early migrants retain cultural heritage practices? How did people adapt their cultural traditions to local environments? What systems were in place to support the continuation of these practices, and how have these systems evolved over the past century? Drawing from data collected in Vancouver, Hong Kong, and Kaiping in rural south China, the project will combine historical document research with virtual reality (VR) technology to produce an immersive VR film about the history of overseas Chinese. TO BE CONT’D

Évaluation de l’analyse hiérarchique de tâche pour déterminer des exigences permettant de concevoir des technologies numériques basées sur l’intelligence artificielle

Les avancées en intelligence artificielle (IA) permettent de plus en plus d’automatiser des tâches hautement cognitives qui, jusqu’à présent, étaient réalisées exclusivement par des humains. Or, la littérature sur la spécification des exigences ne couvre pas spécifiquement les technologies numériques (TN) basées sur l’IA étant donné qu’elle implique un changement de paradigme fondamental en développement de TN.

Le domaine de l’ergonomie cognitive propose un ensemble de méthodes d’analyse de la tâche permettant de définir les exigences cognitives sur lesquels doivent s’appuyer la conception d’environnement de travail. Puisque l’IA consiste essentiellement à automatiser divers processus cognitifs humains, ce projet de recherche vise à évaluer l’application d’une méthode d’analyse de la tâche—l’analyse hiérarchique de la tâche—pour la conception de TN basée sur l’IA.

Eco-Industrial Networking: Developing an Eco-Industrial Network for the Pearson Eco-Business Zone

A collaborative research project between the University of Waterloo, Partners in Project Green (PPG) and the Toronto and Region Conservation authority (TRCA) will attempt to better understand the waste management systems of industrial companies in the surrounding the Pearson Airport so as to assist with goals of urban sustainability in the province of Ontario. Both the TRCA and PPG have a major interest in developing an eco-industrial model in which businesses cooperate to reduce waste and pollution by sharing resources and recycling waste within specific geographic borders. Given these objectives this study will utilize a basic questionnaire to gain knowledge on the waste management systems of the companies in the Pearson eco Zone. The study will identify attitudes and concerns of the companies within the Pearson zone to gain a deeper understanding of how companies can reduce waste and increase overall levels of sustainability contributing to the Eco-industrial model.

Novel self-optimizing algorithm for design and control of power generation and delivery systems

n power generation and delivery systems, there are always a number of control/design parameters and set-points that should be tuned to some optimum values for maximizing efficiency and output power. For example, in photovoltaic cells, there is an optimum load value that maximizes the output power. This value may change over time depending on the PV cell characteristics and environmental conditions. There are different ways of tuning system variables. Sometimes an analytical model of the system under study is used for finding the optimum values. A better alternative is use of intelligent algorithms that play with the controllable parameters to find the optimum point independent of the possible errors in an analytical model. In the proposed research, a novel, high-performance self-optimizing algorithm will be developed. The controller will be applied to famous control problems in renewable energy and power systems. TO BE CONT’D

Optimization of the force sensitive mechanism for football and bicycle helmets

A helmet and its components such as chin strap are designed, tested, and certified for compression force only while neglecting sharp twisting of the brain. The goal of this project is to develop and integrate a Force Sensitive Mechanism (FSM) to a chin strap of a football and bicycle helmet. FSM offers a controlled tightness of the chin strap during impact. FSM can significantly improve the protection of the head against sharp twisting of the brain. In addition, FSM can reduce the risk of neck and throat injuries during an impact. Helmets equipped with FSM will undergo standard helmet tests such as impact test and retention system strength test to ensure compliance and safety

Radiomic-based deep learning for time-to-event outcome in pulmonary malignancies.

Advances in medical image scanning technologies has allowed for great improvement in medical care, from early tumor detection, to trailered treatments and predicting treatment outcome. However, this has resulted in the generation of huge medical image databases, with thousands of medical image scans per patient that need to be examined by clinicians, making it impractical for clinicians to study all the images. This has led to the development of computer-assisted detection programs to automate part of this process. In this project we aim to develop a computer algorithm that would be used for predicting the treatment outcome for lung cancer patient. This project will focus on a specific group of lung cancer patients treated with focused radiation therapy.