In earlier blogs, we explained our collaboration with ResearchConnect to help mutual customers to find relevant funding opportunities. With Impacter we focus on writing better grant proposals, but the language and AI technologies we develop for this purpose can also be relevant in the context of fund searching.
There’s an abundance of funding available in a variety of places such as the EU’s Horizon Europe programme. A significant amount of funds are not utilised, which is a shame considering the amount of talent in universities that want to make a positive impact in the world with their research. Research funding is a crucial step in the innovation process since it funds early stage collaborations between academia and industry and new promising ideas and technology that is not yet worthwhile to invest in for companies and investors. Like Mazzucato argues in her books, funding from the state is crucial in innovation. Therefore, it’s important that researchers in universities can find the right funding instrument for their ideas. Combining the funding database of ResearchConnect with the AI technology of Impacter is a solution to this issue of struggling to find the right funds. And since this summer, this is now available within the ResearchConnect platform for the smoothest user experience.
How it works
In the classic way of searching, one would use keywords or maybe even Boolean queries to find relevant funding opportunities. Last year we already launched the possibility to search for funding opportunities by taking an abstract of your research paper as an input for the search. On top of this we now provide, fully integrated in the ResearchConnect platform, a way of training your own search algorithm. As mentioned in a previous blogpost, we experimented with training your own search algorithm. By launching it in the ResearchConnect platform, it becomes available for all universities in Europe. And it’s really simple to train your own algorithm. Again, we simply start with an abstract of the work of a researcher. This can be a description of their research, or the abstract of a paper, and even several abstracts of different papers that best define this specific researcher. The basic algorithm will calculate a match, and the training start by you telling the system whether you like (thumbs up) or dislike (thumbs down) the match. Every time you tell the system what you think of the match, it re-evaluates your preferences and changes the search a little. This way you are training your own search model, while at the same time finding relevant funding matches. You can save the model for a better starting point during the next search and you can train as many models as you want!
Curious to learn what this new module can do for your research institution? Reach out to ResearchConnect for a demonstration!