‘Un’natural language processing
SenseiBot is now able to do things like the above. It took another string ‘tunnel junction for reading all four bases with high discrimination’ and correctly labeled it ‘Semiconductors’ and ‘Genetics’. Some people call it natural language processing, but who naturally says things like this!
We’re doing this because we have our hands on the largest collection of historical research data. By understanding the skills that go into each research project, SenseiBot identifies pools of expertise, and so makes better connections.
This guy, who knows a thing or two about all this, suggested ‘topic modeling’ to not be unfair to the rest of us who do speak naturally!