A different take on data skepticism by Beau Cronin.
From the post:
Recently, the Mathbabe (aka Cathy O’Neil) vented some frustration about the pitfalls in applying even simple machine learning (ML) methods like k-nearest neighbors. As data science is democratized, she worries that naive practitioners will shoot themselves in the foot because these tools can offer very misleading results. Maybe data science is best left to the pros? Mike Loukides picked up this thread, calling for healthy skepticism in our approach to data and implicitly cautioning against a “cargo cult” approach in which data collection and analysis methods are blindly copied from previous efforts without sufficient attempts to understand their potential biases and shortcomings.
…Well, I would argue that all ML methods are not created equal with regard to their safety. In fact, it is exactly some of the simplest (and most widely used) methods that are the most dangerous.
Why? Because these methods have lots of hidden assumptions. Well, maybe the assumptions aren’t so much hidden as nodded-at-but-rarely-questioned. A good analogy might be jumping to the sentencing phase of a criminal trial without first assessing guilt: asking “What is the punishment that best fits this crime?” before asking “Did the defendant actually commit a crime? And if so, which one?” As another example of a simple-yet-dangerous method, k-means clustering assumes a value for k, the number of clusters, even though there may not be a “good” way to divide the data into this many buckets. Maybe seven buckets provides a much more natural explanation than four. Or maybe the data, as observed, is truly undifferentiated and any effort to split it up will result in arbitrary and misleading distinctions. Shouldn’t our methods ask these more fundamental questions as well?
Beau make several good points on questioning data methods.
I would extend those “…more fundamental questions…” to data as well.
Data, at least as far as I know, doesn’t drop from the sky. It is collected, generated, sometimes both, by design.
That design had some reason for collecting that data, in some particular way and in a given format.
Like methods, data stands mute with regard to those designs, what choices were made, by who and for what reason?
Giving voice what can be known about methods and data falls to human users.