Innovative Projects Realized

Behavioral mapping Using Machine Learning and Natural Language Processing to Predict Individual Behaviors and Team Dynamics

This project tackles the issue of knowledge incompleteness and lack of domain coverage in resume and job posting matching caused by the exploitation of domain-general resources. A variety of co-operative semantic/ontological resources will be used to filter out irrelevant resumes. A two-way (candidate to job and job to candidate) semantic-based automatic suitability ranking is proposed. The suitability is determined by the semantic distance of resumes and job postings, evaluated by their word embeddings. An efficient semantic space created through the Convolutional and Recurrent Neural Networks will be utilized with different word embedding mechanisms along with different classification methods. The project also investigates the potential of knowledge graphs in illustrating inconsistencies between the resumes and job postings will be investigated. This study improves an automatic system capable of precisely detecting, extracting, and visualizing the resume and job posting’s relevant skills as well as the implicitly encoded semantic dimensions of applicant resumes.

Improvements to the Environment Footprint Monitoring Platform

The EFMP is an environmental management platform. At its basic level it monitoring the environmental conditions over time and over the entire footprint of a project. The footprint of a development does not end at the boundary fence – the area that the development effects on a regular basis. The environmental information comes from Earth Observation Satellites, ground-based sensors, and UAVs. The platform integrates the data and presents it in a form that managers can use. Our goal is to improve the existing platform and to add more functionality in the areas of incorporating internal and external models, This requires careful design of new associated data tables and formats.

Détermination de l’habitat d’origine chez le Pic à dos noir Picoides arcticus et de Pic à dos rayé P. dorsalis en dispersion juvénile

Ce projet novateur vise à mieux comprendre la dispersion juvénile chez le Pic à dos noir et le Pic à dos rayé, deux pic bois de la forêt boréale de conifères. Dans cette recherche, on utilise les composés chimiques présents à la surface des troncs d’arbres et qui se déposent sur les plumes de la queue des pics lorsqu’ils s’alimentent. À partir d’adultes nicheurs occupant des habitats homogènes d’une essence forestière, nous identifions les composés qui caractérisent les différents habitats dans lesquels ces pics évoluent. Ces informations permettront de produire la “pierre de rosette-la signature chimique” qui servira à déterminer les habitats d’origine des juvéniles qui se dispersent à l’automne et qui sont recensées par l’Observatoire d’oiseaux de Tadoussac (OOT). Ce projet permettra de maximiser les travaux de l’OOT en bonifiant l’interprétation toucahnt les mouvements de ces oiseaux emblématiques de la forêt boréale.

Étude du stockage thermique pour supporter la transition énergétique

Ce projet permettra de gérer la forte demande énergétique sur le réseau électrique grâce à l’utilisation de technologies de stockage énergétique dont les accumulateurs thermiques et le stockage à air comprimé. De façon plus spécifique, le projet a pour objectif de développer un outil de design pour l’implantation efficiente des accumulateurs thermiques et de tracer les grandes lignes pour leur couplage efficace à la géothermie. De plus, le développement et la commercialisation des accumulateurs thermiques seront simplifiés grâce aux outils développés dans le cadre de ce projet, ce qui favorisera la transition énergétique. Enfin, ce projet permettra également d’assurer la formation de la relève dans le domaine de la transition énergétique.

Forcasting the Capabilities of LIM-pair Estimartors to Probe The Cosmic Dark Ages

While we have been able to learn a great deal about the early moments after the Big Bang from cosmic microwave background (CMB) experiments, most of the universe remains unexplored. In particular, the cosmic Dark Ages, a time before any stars or galaxies existed, remains a mystery. This is a time when the cosmos was filled almost entirely with neutral light atoms. The distribution of this matter, though seemingly random, has encoded in it deep quantum structure that originated before the Big Bang and that gave rise to all of the complex galactic physics we see today. A peek into the cosmic Dark Ages would give us a clear view into the beginnings of our universe. This, however, is no small feat. Between us and the early universe are billions of years of comic evolution and in looking deep into the cosmos, our view is contaminated by late-time cosmology. But, there may be a way out.

Development and evaluation of a made-in-Manitoba data rover for the rapid collection of ground-level phenotypic data

Global food security is one of the most pressing issues of our time. Developing new farming practices and technologies that take advantage of large amounts of data collected from a myriad of different sensors offers our best bet for a greener future, with a stable supply of nutritious food for everyone. A move toward digital agriculture, as it is called, is expected to bring critical innovations such as automated methods of plant care and inspection. Such capabilities will increase crop yields and expedite breeding programs, while reducing human effort, costs, and chemical inputs (e.g., fertilizer, herbicide, pesticide). The proposed project will help bring out these innovations by developing a data rover that will collect plant level images and environmental data in test plots. This will provide much greater information about plant growth, resilience, and yield (i.e. phenotypic data) than can presently be collected by hand, allowing plant breeders to make better decisions and develop crops with improved characteristics (e.g. drought resistant) more quickly.

Financial Governance and Modelling for Coastal Shellfish Corporation Strategic Plan

The purpose of this project is to provide Coastal Shellfish Corporation with an integrated biological and financial modelling framework for the management of current and future aquaculture operations of the corporation. The primary goal of this research is to provide a framework for implementation to the management of CSC that will enable to organization to implement best practices for biological and financial modelling for coastal shellfish aquaculture at farm-scale level.

Conveniently Green – Market Development and Expansion – BSI Application

Sales Strategies
• Researching, analyzing, and testing new communication and sales strategies that align appropriately with our target market
• Developing innovative ways to manage our customer relationships efficiently and with a high level of organization by researching customer relationship software in which CG can implement.

Marketing Strategies
• Develop and implement effective social media content and advertisements to drive foot traffic to current retail customers and promote CG products in potential client businesses.
• Research and analyze potential digital marketing campaigns designed to promote CG products through internet platforms (Facebook Ads, Google Ads, and search engine optimization)

New Product Research
• Research new products identified as being correlated with CG’s target market, including identifying suppliers and manufacturers.
• Create forecasting models to predict sales volumes, predicted margins and forecasted revenue.

NLP-empowered Automation for Integrity Assessment in Legal Documents

Natural Language Processing (NLP) is a learning-based technique under the general scope of Artificial Intelligence (AI). NLP has shown its benefits in many application, including finance, accounting and auditing. Differing from semantic extraction in social
media, text interpretation for auditing has the legal constraints, which is crucial in determining whether an invoice or a contract is legitimate from the legal perspective. Legal documents can range from a few pages to hundreds of pages, which is time
consuming for domain experts to review. Our goal is to automate such process using NLP techniques and AI algorithms to improve time and accuracy, avoiding manual errors and replacing human procedural checking, and thus releasing the valuable time of auditors to focus on critical issues.

Les meilleures pratiques à portée de main: Soutien au développement d’un programme de formation continue multi-composantes pour et par le Réseau Avant de Craquer

Les proches de personnes ayant un trouble de santé mentale peuvent obtenir du soutien dans les associations de leur communauté. À ce jour, le milieu communautaire n’a pas défini quelles sont les meilleures pratiques spécifiques à leur mandat qui est d’offrir du soutien psychosocial communautaire aux proches en santé mentale et ce, en fonction des nombreuses caractéristiques de la clientèle (âges, liens avec la personne ayant un trouble mental, types de trouble, etc.). De plus, les intervenants de ces associations sont souvent isolés et n’ont pas les ressources pour développer leurs pratiques. Le projet vise à définir les meilleures pratiques de soutien psychosocial communautaire et à développer un programme de formation continue afin de favoriser la qualité et l’amélioration continue des services offerts par ces intervenants. Enfin, le projet contribuera à la reconnaissance de l’importante contribution des organismes communautaires dans le continuum de services en santé mentale.

Design and development of a multi-channel automated Hipot and continuity tester for medical devices

A novel medical device for the treatment of heart arrythmia is designed and built by the partner organization. The medical device must go under certain electrical safety tests in the final inspection. These tests are currently conducted manually by the operators. This stage is a major bottleneck in the production line. On the other hand, the device is certified to be used in Europe and will start a controlled Clinical study in the USA under a PMA filling. Because of this, the partner organization wants the intern to undertake a research project whose objective is to automate electrical safety tests and so decrees the testing time. The main benefit to the company is increased number of devices that pass final inspection, enabling the company to make the supply meet the high demand in the future, and more importantly having more patients being treated by the device.

Proximate mechanisms behind carry-over effects mediated by key reproductive hormones in migratory seabirds

Long-lived animals, including seabirds, must balance investment in current reproduction with investment towards survival and future reproduction to maximize their lifetime fitness. Breeding is a critical stage in birds annual life cycle and requires high energy expenditure. Increased reproductive effort in the current year can therefore lead to costs that will have downstream consequences into subsequent life stages, including migration and overwintering or the next breeding season. Such carry-over effects can have strong impacts on fitness and population dynamics, but mechanisms underlying such seasonal interactions remain unclear. I monitored breeding and migration of an Arctic-breeding seabird, the black-legged kittiwake (Rissa tridactyla), and captured individuals at their return to the colony site to assess the physiological state and reproductive preparedness of individuals as they transition into the next breeding season. We used hormone challenges and collected blood samples to assess levels of the ‘stress hormone’ corticosterone and key reproductive hormones: luteinizing, estrogen and testosterone. Hormones, and endocrine processes, are the mediators linking the individual perception of its environment and its reproductive decisions and success, driving the ability of individuals to cope with environmental stressors.