Skillicorn's 499 Projects, 2017-2018
Here are some projects I am willing to supervise.
I'm part of a group looking at detecting outcomes in leukemia based datasets about patient genotype and phenotype. Although prediction accuracy has increased with newer predictive techniques, there are still some open issues. Using gradient boosted predictors, and other newer techniques, may help. (Taken)
I was part of a group that discovered that detecting fraudulent MD&A company filings was easier using their textual content than their financial details. New prediction techniques might change this. The project is to revisit fraud prediction in both text and financials, using leading edge techniques.
One way to change the advantage attackers have over defenders in cyberspace is to create and manage many plausible-looking but fake versions of real documents. Someone trying to exfiltrate information doesn't know which is the real version, but the system itself can keep track straightforwardly. This project is to build a proof of concept implementation of such a file system.
I am also willing to supervise other projects that are related to my research interests. If you have an idea, please suggest it.
All of these projects could be preparatory work for Master's theses if you want to begin graduate study in May 2018 (finishing a year later).
Back to David Skillicorn's home page.