Tamfang

Undoing a series of disruptive edits edit

Latest comment: 1 month ago2 comments2 people in discussion

You asked about undoing a series of bad edits. This is how

Use the View history tab to identify the last good version and the last of the disruptive edits, then Compare versions
Use either the [restore this version] on the left pane or the undo on the right.

If editing on mobile it is a bit more painful but essentially you choose the history tab too but then you select the last good version and do a null edit on that.

If you need to do it a lot, you might want to request WP:ROLLBACK privilege. I've never bothered. 𝕁𝕄𝔽 (talk) 17:47, 27 March 2024 (UTC)Reply

I was given Rollback long ago (unrequested), but the series in question did not include the last edit, so no rollback. —Tamfang (talk) 17:51, 27 March 2024 (UTC)Reply

Predictions in statistics edit

Latest comment: 1 day ago2 comments2 people in discussion

Hi, I just wanted to follow up with you about your modification of my edit on the "Cat" page. I see no reason to challenge it, as it might be clearer for some people. However, "predict" here was not meant to say "to detect later pain". It referred to predictive validity. When a score on a test or scale correlates well with the outcome of interest, as later measured in some other way, we can say it "predicts" it. Concerning the Feline Grimace Scale, this paper uses the term: https://www.nature.com/articles/s41598-023-49031-2 Here is an excerpt from the abstract: "This study used deep neural networks and machine learning models to predict facial landmark positions and pain scores using the Feline Grimace Scale© (FGS)." The authors don't mean here that they're trying to predict what facial landmark positions will show up at a later time, but rather that they want their deep neural network to describe correctly what the grimace scale will find. In that way, it "predicts" what the scale would find. Here is another example, this time about depression and anxiety: https://www.nature.com/articles/s41598-021-81368-4 The authors don't try to predict future depression or anxiety, but what the tests measuring related symptoms will find (or in this case, have already found).

While in this case your word choice seems fine to me, I'm writing you this as I thought it might be useful for future edits, where changing the word "predict" might in fact alter the meaning of what is said.

All the best! Leafy Gee (talk) 01:44, 5 May 2024 (UTC)Reply

Okay. I did understand that possible intent, but even so I thought it not best to spring on a lay audience without supporting context. —Tamfang (talk) 07:11, 5 May 2024 (UTC)Reply

Add topic