Machine learning is too often perceived as a kind of oracle-like power capable of predicting the unknowable and working out of distribution.
Show this thread
-
But the “proper setup” would be using a less useful target variable (i.e., article is predicted to be evergreen rather than article actually withstood the test of time). So there’s a tradeoff between these two possible setups, right?
There's no tradeoff here. You have to understand that such classifiers work not because they have a crystal ball, but because they latch onto superficial markers that are statistically correlated with evergreeness or not (e.g. lots of dates = not evergreen).
-
-
The difference in ratings between what people could predict today (wrt current articles) and what people would think 20 years from now is not something that a classifier could possibly predict. So nothing is lost.
-
At the same time, the classifier trained on old articles would *simply not work* on today's articles. Too many words/concepts would be unknown or redefined, and the style cues it uses would be outdated. It would produce largely garbage output (or at least very biased)
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.