New NSERC-CSE Research Grant Available and it's Very Interesting
The NSERC-CSE Research Communities Grants has a new research grant for unstructured data analysis
The Natural Sciences and Engineering Research Council of Canada (NSERC) and the Communications Security Establishment (CSE) is a joint partnership to strategically fund cutting-edge technology that is of importance to CSE and the Government of Canada.
The NSERC-CSE Research Communities grants offer two types of grants based on the size of the project in question:
$700,000 per year over four years for small projects
$1,400,000 per year over four years for large projects
This grant is primarily for universities, and CSE are encouraging inter-university collaboration. Despite this focused on academia, there is a lot of potential for partnerships with industry. There will be Top Secret components to this research as well, which raises a lot of interesting questions about what CSE aims to accomplish with this grant.
What this actually could mean…
Unstructured data, as the name applies, is data that is not organized in any meaningful or useful manner. In a very rudimentary sense, unstructured data is a massive list of numbers and data that has not been added to an Excel sheet. Once it has been added to an Excel sheet, it becomes structured. Now, broaden this concept to consider all data on the Internet as unstructured in some manner. These days, structured data is all the more important because of the significant acceleration of artificial intelligence (AI) and machine learning (ML). While data has been important for decades, clean, structured data, in particular, is of monumental importance to train AI/ML models. Good data is needed to train models, whether it is a large language model or specialized models, to identify air threats.
It is enormously telling that this research grant mentions algorithms and ML, but does not mention AI. It is clear that CSE is approaching this with a very specific use in mind and is hoping to avoid the FUD that exists in a lot of AI product marketing at the moment. What CSE hopes to fund with this grant could have massive potential for CSE beyond the ability to acquire good data for AI/ML models, as the ability to use unstructured data is not solely about turning it into structured data but establishing methods of analysis that don’t require the data to be structured.
This will be a project to watch and see what is publicly released.