Research into Convolutional Neural Network (CNN) Explainability

Machine Learning is advancing at an astounding rate. It is powered by complex models such as deep neural networks (DNNs). These models have a wide range of real-world applications, in fields like Computer Vision, Natural Language Processing, Information Retrieval and others. But Machine Learning is not without some serious limitations and drawbacks. The most serious one is the lack of transparency in their inferences, which works against relying completely in these models and leaves users with little understanding of how particular decisions are made.

Assessing and managing acoustic disturbance to bowhead whales in the Canadian Arctic

Scientists of WCS Canada have obtained funding through the Canada Nature Fund for Aquatic Species at Risk (CNFASAR) to conduct a collaborative project focusing on bowhead whale research in the Canadian Arctic. The proposed postdoctoral project is a main part of the CNFASAR project and aims to assess how bowhead whales react to underwater noise so that risks from human activities, particularly ship-related, can be managed effectively.

Automobile Purchasing Behaviorial Data Collection, Management, and Analysis

This project investigates automobile purchasing behavior of female millennials. In order to achieve the goal of understanding and making use of purchasing behavior, data are to be collected, managed, and analyzed. In addition to using existing data and third-party data, two major tasks of data collection are the use of questionnaire and web crawling to gather region, product types, and consumer market segment specific information relevant to the partner organization.

Synthesis of graphene and graphitic films

The overall problem to be addressed is the synthesis of graphene or graphitic coatings from the liquid phase. This work follows from a previous investigation, supported by a MITACS Accelerate grant, into the use of “poly(hydridocarbyne)” (PHC), a soluble carbon- hydrogen polymer, as a precursor for the formation of diamond- like carbon coatings. As a result of attempts to purify PHC by electrochemical cycling it was discovered that the surface of the polymer could be converted into a graphene or graphitic layer.

Photocatalytic Oxidation of Volatile Organic Compounds in Air

A new type of device that uses a combination of UV light, oxygen, flowing water, and titanium dioxide (a com-mon white pigment) is capable of removing volatile toxins from the air. This device will be investigated using a range of techniques that provide molecular insights into this process, and those insights will help in the rede-sign of the device to make it as effective as possible, in terms of energy efficiency, longevity, reusability, range of compounds removed, and speed of their removal.

Winter Ecology of Chinook Salmon in the Canadian Salish Sea

Chinook Salmon are a species of high ecological, economic and cultural value in BC. Recent declines in Chinook Salmon abundance have highlighted a need to understand factors controlling their productivity. One hypothesis suggests that the first winter in the ocean plays a critical role in controlling Chinook Salmon survival, and in turn, abundance. Little research has been conducted during the winter, limiting our understanding of this potentially critical period.

Achieving consistently flavoured sour beers through better chemical understanding

The popularity of sour beers is continuously increasing. Producing sour beers is time consuming and obtaining a consistent flavor profile over multiple batches can be challenging. This in addition to scaling up production to meet customer demands can negatively influence the quality and flavor of the beer. This project aims to develop advanced analytical techniques to help understand the relationship between chemical composition and flavor.

The effects of grapevine microclimate manipulations on Pinot Gris glycoside aroma compound abundance and diversity

This research project is a partnership between The Wine Islands Grower’s Association, and the University of Victoria. The goal of the project is to better understand how vineyard management techniques effect wine grape quality prior to wine processing. Pinot gris is the most important white wine varietal grown on Vancouver Island and throughout British Columbia. To understand how vineyard techniques affect quality gas chromatography and mass spectrometry will be used to identify and qualify the aromatic compounds present at harvest to maximize the potential of the grapes.

Indirect effects of predator control: Examining predator habitat selection and competitive interactions following wolf control in northeastern Alberta

Global demand for natural resources is resulting in unprecedented landscape change. In northeastern Alberta, woodland caribou persistence is threatened by habitat loss associated with natural resource extraction, and increased predaiion by grey wolves as a result, due to their ability to capitalize upon landscape disturbance via increased movement on linear features. Wolf control via culling is a common wildlife management strategy to conserve woodland caribou.

Using wearable sensor-based technologies to detect changes in health status for prevention of adverse health events and to improve overall quality of life - Year two

The project goal is to determine the clinical utility of Orpyx LogR technology to detect gait changes and their efficacy to predict and monitor fall risk. Project I will use existing data to determine sensitivity and specificity for prospective classification of fallers and non-fallers for a composite measure drawn from an extensive battery including single and/or dual-task IMU-derived gait metrics as well as from force plate gait initiation data.