Adaptive Visualization for Analysis of Customer Behaviour

Companies with a strong online presence are seeking to increase users’ engagement by adding recommendation and prediction algorithms to their websites. These algorithms can for example predict which users’ actions are more likely to happen and which users’ actions are likely more profitable. Companies use such algorithms to tailor the content of their websites based on the users’ behavior. Websites tailored to users’ behavoiur become more profitable and competitive for digital advertisement and e-commerce.

Software Engineering for Mobile Game Software Product Lines

This project investigates the cost effectiveness of applying product line techniques to the development of the common elements of certain types of video games. Approximately ten games will be analyzed to identify those parts that are used by most of the games. Those commonly used parts will then be re-engineered to make them more generic and to facilitate their reuse in new games, thereby decreasing the cost of production and reducing the time it takes to bring a new video game to market.

Targeting achievement in higher-order thinking and STEM: An interprofessional approach

This proposed research investigates the use of ThUMP by classroom educators to foster high order thinking through authentic and meaningful practice activities. Research suggests that practice is require to obtain expertise in complex subject areas such as STEM (Science, Technology, Engineering and Mathematics). ThUMP allows educators to author practice activities that students can access using mobile devices. This proposed research adopts a design based approach in order to provide feedback / feed forward to the Mathtoon’s developers.


ProStudioMasters is poised to become a key player in the emerging market of high quality, high bandwidth, uncompressed audio distributed over the internet. The interns will provide support by compiling information to be included in the container file along with the audio: metadata such as correct attributions (composers, performers), historical circumstances of the recording, analytic capsule summaries of classical works, aesthetic and cultural importance of the recording, and other necessary items that vary with the repertoire.

Multi-Category Classification Confidence for Ad Contextualization

This project studies Machine Learning algorithms for multi-category document classification. The purpose is to effectively predict user’s behavior based on the contextualization of the advertising and the associated document and therefore, to increase the click rate and the success of a dynamical advertising campaign. Due to the nature of the World Wide Web, the feature sets for the classification is extremely large. However, many learning algorithms don’t perform well with large number of features or attributes.

Middleware infrastructure for processing of big spatial data on Spark

Location-based services such as Recon Instruments’ Engage web application provide feedback to users based on analysis of GPS trajectory data. While existing data analysis platforms such as Hadoop- GIS provide simplified support for batch processing of massive amounts of “big spatial data”, they are not efficient for supporting iterative machine learning algorithms (such as those in use at Recon) or interactive queries (common in web applications).

The Smile Epidemic: Investigating the Impact of a Digital Social Media Gratitude Intervention

Research shows that keeping a gratitude journal can enhance well-being. The Smile Epidemic Inc. is a start-up company that has developed a digital, photographic social media update on the traditional (private, written) gratitude journal. The Smile Epidemic invites people to express in a few words what made them happy or grateful that day, and to take a photo of that message. People can them post the photos in an online community (see and share via other social media (e.g., facebook).

Fabrication and Testing of a Microlens Array Optical System

The purpose of this project is to design and fabricate a compact optical component that will be used in tandem with a display panel to use in Head-Up Display systems. The essence of the optical component design is the use of microlens arrays. Microlens arrays are thin, usually planar lenses composed of much smaller lenslets with diameters usually in the order of 100-200 microns. Using microlens arrays, the optical system can be reduced to a couple of “sheets” of microlens arrays or less, allowing for compactness of the system.

Compressive Superresolution Projector

Research into so-called computational displays has shown that new and extended imaging modes can be made possible by augmenting novel optical design with computation. Examples include highdynamic range display via local LED backlights and multilayered autostereoscopic (glasses free 3D) displays. While there has been considerable progress in standard television form-factors, this line of research has only recently started to be explored for projectors.

Using machine learning methods to improve image suggestion and image retrieval results

Vidigami, the sponsor company, is developing tools to create digital yearbooks for schools – organizing personal photo collections of students, parents, and staff. Current technologies have made it easy for everybody to have a digital camera or simply digitalize their hardcopy versions of their photos. However, this huge amount of data makes it hard to sort photos and extract an album based on specific criteria. The problem of managing photos becomes more complicated by the growth of social networks where people can share and see each other’s photos of an event.