Talk:False positives and false negatives

Latest comment: 9 months ago by Emege2023 in topic Usage with opposite meaning

"resp." meaning? edit

Can someone replace the abbreviation "resp." that is currently in the lead?

I'm making the request simply because I don't understand the abbreviation. I am ignorant about the subject, but of course that's the audience that the article should help, right?

Thanks in advance. — Preceding unsigned comment added by DavidHolmes0 (talkcontribs) 14:38, 6 July 2015 (UTC)Reply

  DonePaulTanenbaum (talk) 21:15, 1 September 2015 (UTC)Reply

False elarm edit

A "False negative error" should be called as "False alarm" not the "False positive error". I think there is a mistake. CoperNick (talk) 17:23, 20 February 2017 (UTC)Reply

No, it's correct as is. The alarm is a positive conclusion that the condition exists. Loraof (talk) 18:36, 16 August 2017 (UTC)Reply

Wrong lead section of this article edit

This article explains "False positives and false negatives" in various areas but the lead section explains only meaning in medical statistics using statistics jargon. More common use of the term is explained more accessibly only further in the article. -- pabouk (talk) 11:49, 8 June 2018 (UTC)Reply

I have reverted the lead section to a variant from 2016-04-24T16:19:55 which describes the terms in multiple areas instead of medical only. Probably it is too long and needs some shortening. pabouk (talk) 12:14, 8 June 2018 (UTC)Reply
I haven't checked how much the article has been edited since then but I have tried to simplify the lead section again by removing the various caveats and repetition. PeepleLikeYou (talk) 10:55, 15 August 2020 (UTC)Reply

Question edit

What is thirteen minus 6? Sarfarazhemat (talk) 16:42, 1 February 2019 (UTC)Reply

Is it possible to add the layman description? - "Expected result : Negative, actual result : Positive - then, its called as False Positive - i.e., when negative is not negative, its false positive" — Preceding unsigned comment added by Bbee1915 (talkcontribs) 12:02, 20 March 2019 (UTC)Reply

Bad Examples edit

I came across this article, and noticed that some of the examples in it are rather poor. The article uses "The Boy Who Cried Wolf" as an example of a false positive, but that story involves a person deliberately lying. False positives are not about fraud, but about the parameters of a test. I can't think of a good example off the top of my head, but I think a straightforward example from medicine could be the best option. The "Consequences" section is also strange, because it talks about the impacts of false convictions and acquittals in court. These scenarios are not usually called "false positives/negatives," nor are they mentioned elsewhere in the article. I was thinking that the section should be renamed "Impact" and rewritten from scratch without using any of the existing content. It should discuss how a test results are confirmed in medicine and how false results are detected, as well as what effect false positives have on statistical analysis and correlation. Does this approach sound good to others? --DavidK93 (talk) 15:53, 17 December 2019 (UTC)Reply

@User:DavidK93 The example of the boy who cried wolf is very confusing for the reader, as you say. But the pregnancy test or the guilty/not guilty scenario are actually good examples of false negatives and false positives. I've rewritten it, let me know what you think. PeepleLikeYou (talk) 09:28, 15 August 2020 (UTC)Reply

Image edit

 

I created a diagram on PowerPoint about this topic, but not sure if this would be sensible enough for it. Qwertyxp2000 (talk | contribs) 03:10, 2 August 2022 (UTC)Reply

Usage with opposite meaning edit

In the context of Formal Verification of Hardware systems, a "Pass" result (the property is proven, no counter-example was found) has been usually described as a "positive" result (and a "Fail" result, a "negative"). In that context, procedures that yield invalid "pass" results are said to produce "false positives", and those that find potential counter-examples that are not real are said to produce "false negatives".

For example:

1) https://www.oreilly.com/library/view/formal-verification/9780128008157/xhtml/chp009.xhtml

"... cases where FV techniques were applied incorrectly, sometimes resulting in false positives: cases where a “formally proven” design turned out to contain errors ..."

2) http://what-when-how.com/Tutorial/topic-3571uclo1c/SVA-The-Power-of-Assertions-in-SystemVerilog-458.html

"... either a false negative, in which case failure is reported for an assertion that should pass, or a false positive, in which case success is reported for an assertion that should fail ... " Emege2023 (talk) 21:16, 17 July 2023 (UTC)Reply