This is a reply I wrote in 2017 as part of an interesting discussion with a colleague concerning whether idealism, or anti-materialism, had any place in a naturalistic philosophy of science.
Even without going as far as scepticism surrounding causation or induction, there are plenty of problems if we consider the body of science to merely be a vast collection of perceptions. My main point of interest here – and one which I think has real relevance to data science – is the different between “knowledge” and “understanding”.
But I wouldn’t be too quick to write off Berkeley’s idealism (or anti-materialism, if you will).