I’m trying to help scientists do science

duma2Would you like to get more insight in the world of text and data miners? Daniel Duma is a PhD student at the Alan Turing Institute and the University of Edinburgh and he shares his story in a short movie. He is working on software that will recommend relevant papers to scientists writing papers.

 “What I’m working on is recommending relevant papers to people writing papers. So I’m trying to help scientists find the relevant science on which to base their research. With which to backup their claims on which to fundament their approaches. So basically I’m trying to help scientists do science.


So the end result hopefully, will be sort of like a software, that you can plug into your existing Word processor or text editor. It will recommend papers that you should be aware of, you should read, or that you would want to cite, based on what you’re saying.

So I’m trying to apply some measure of understanding to the text that you’re writing, and to the text that has been written before, in order to connect the two of them. To say, ok you don’t know of this author, but they wrote a paper where they were saying something that is very similar to what you are saying in your paper, so maybe you should go and read that paper, because maybe that is relevant to your work.  This is what I’m trying to build.

I would like for this to be an open academic project. I would like people to use it, I would like it to be a tool that people can use, and I would like them to collaborate, to contribute to it. I want it to be open source, only using open access publications. This is important because it’s an absolute nightmare to deal with publishers in terms of accessing their copyrighted material, to index it even.

I think there is a lot to be optimistic about and to be encouraged by, in the field of data mining currently, or text and data mining let’s say. The applications with this are infinite, I mean, you can do anything. You can eventually make robot scientists. You can eventually get algorithms to help you in your research at a human level, you know.

I think this is the future to come, it’s not that far away. And I’m excited to work on it, and I encourage other people to join and work on this as well.”

Related links: