Improved Commentary Prediction on Financial Data

Companies rely on financial reports which are generated through various transactions such as sales and expenses to understand the discrepancies between actual performance and financial forecast. Accordingly, generating commentaries on financial data might be considered as a routine operation for many companies. The previous studies indicate that machine learning algorithms can be used to automate the process of commentary generation. Specifically, such approaches use product forecasts and actuals in addition to inventory and point-of-sales data for the underlying prediction task.

Smart Dashboard for Sustainable Destination Decision Making

Destinations have quickly become victims of their own success. Destination Management Organisations (DMO’s) worldwide are making a much needed shift towards the inclusion of management alongside their marketing priorities for destination management, but are often ill equipped. There is a current gap in the marketplace for useful, comprehensive and user-friendly tools to assist them.

Machine Learning for Increasing the Speed and Quality of Service in Online Medical Services

Telehealth/Telemedicine covers a variety of medical services offered through media such as telephone, email and the internet. As technology evolves, these online medical services can be further enhanced to improve the user experience. In this project, we aim to focus on a particular set-up where the patients interact with doctors over a chat service. The chat service employs a number of doctors with varying specialties, and at any time during a doctor’s work hour, he/she might interact with up to five patients concurrently.

Predicting the Fair Market Value of a Real Estate Asset

Finding the right real estate property that grows in value over the next few years is of paramount importance. For doing so, one of the most important factors is to estimate the current value of the property as well as its future value. The goal of this project is to build a data and domain-driven model using machine learning that uses previous real estate data to estimate the value of property and suggest the right property in the right neighborhood for investment. We will build an end-to-end framework to collect and preprocess the data and then predict the value of a property.

The use of an affect-based music selection algorithm and embedded auditory beat stimulation as an intervention for high-anxiety populations

Chronic anxiety is a growing psychological disorder worldwide and in Canada. Even when anxiety presents at pre-clinical levels, it can be disabling. Anti-anxiety drugs have many adverse side-effects. In some cases, listening to music decreases anxiety more effectively than anti-anxiety drugs. Furthermore, some evidence suggests that curating a music selection as a function of the listener’s current affective state allows for more effective and sustained changes in mood. Auditory beat stimulation (ABS) may also reduce anxiety and promote relaxation.

Design of Novel Catheter Systems for Local Drug Delivery and Cryogenic Lung Biopsy

Balloons have potential benefits over stents for localised drug delivery. The activities in this proposal will aid the development of a dual balloon catheter system. By the end of the internship, the balloon catheter design will be completed and frozen in anticipation of pre-clinical studies.
The gold standard for diagnosis of interstitial lung disease is surgical biopsy, despite associated complications including mortality. Cryogenic lung biopsy presents a new method of diagnosis with low risks compared to surgical methods.

Enzymatic nano-immobilisation facilitated by 2D materials for antifouling coatings - Year two

Enzyme immobilisation is crucial for preserving the enzyme activity while enabling the enzymes to be recovered and reused for multiple applications in biocatalysis. However, immobilisation can change the structure and functionality of enzymes. Therefore, immobilisation of enzymes needs to be carefully investigated and controlled at fundamental levels. The emerging two-dimensional (2D) materials, such as graphene and transition metal dichalcogenides exhibit unique physico-chemical properties which make them well suited for enzyme immobilisation.

Methods to improve SNR of a small-scale NMR system forin-vivo biomarkers monitoring

The goal of this project is to develop advanced mathematical algorithms integrated with the probe
design technique to improve the SNR of an NMR measurement. The proposed methods are summarized as:
1) Apply Compressed sensing (CS) technique associated with the optimized recovery method to the current obtained NMR signals.
2) Time-averaging technique can also be developed and applied to further improve the SNR of the developed algorithms.
3) Investigate and develop the phased-array receive (Rx) coils for the current MR system.

Testing, Integration, and Optimal Control Strategy of Residential Hybrid HVAC System

The Canadian federal government committed to encouraging low carbon alternatives and the growth of clean technology that reduces greenhouse gas (GHG) emissions. It is stated that the new target is to reduce GHG emissions by 80% by 2050, relative to 2005 GHG levels. In order to achieve this goal, one of the government’s strategic plan is to promote systems and technologies that minimize natural gas/fossil fuel usage and increase the use of clean electricity.

Air Quality and Heat-Related Health (and Death) Effects of Increasing Green areas: the case study of the Greenbelt’s Urban River Valleys

The Greater Toronto Area has experienced significant urbanization during the past decade; meanwhile, the consequences of the urban heat island and the frequency and duration of the heat waves are becoming more evident. Preserving the green areas and increasing vegetation leads to a decrease in air temperature, an increase in evapotranspiration, a decrease in cooling energy demand; and provide better thermal comfort for inhabitants.