Talk:Ground truth

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Meteorology (Rated Start-class, Low-importance)
WikiProject icon This article is within the scope of WikiProject Meteorology, a collaborative effort to improve the coverage of Meteorology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
Start-Class article Start  This article has been rated as Start-Class on the project's quality scale.
 Low  This article has been rated as Low-importance on the project's importance scale.
 

Ground truthing in machine learning[edit]

What is "objective data"? --MartinThoma (talk) 17:11, 23 July 2014 (UTC)

Objective data can be proven. Contrast with subjective data, which is driven by opinion. I'll add that to the page. Adam KatzΔ 03:37, 30 January 2015 (UTC)

Ground truth vs. Gold standard[edit]

As noted in Talk:Gold standard (test), the term "gold standard" is actually more faithful than "ground truth" with respect to most usage in machine learning and information retrieval. This is because ML and IR make assumptions about the classification of labeled data (mostly because human classification, which is often itself error-prone, does not scale). I had to resist "correcting" a recent edit that muddied this point from the ML perspective. That said, these and related fields of study do not (in my experience) use the term "gold standard," even if it is more appropriate. Adam KatzΔ 20:07, 8 January 2016 (UTC)

All definitions are identical[edit]

As far as I can tell, these definitions are all identical, with very minor differences. However, I'm only expert on the machine learning part, so I'm not comfortable doing that myself. Adam KatzΔ 00:02, 27 February 2015 (UTC)

Confusion about paragraph of GIS[edit]

I do not like how the paragraph "Geographical Information Systems" begins:

Geographic information systems such as GIS, GPS, and GNSS,

GIS is just synonym for Geographic information system. And how is GPS a GIS?? This should be re-worded, but I am not sure how. Rmasta (talk) —Preceding undated comment added 09:41, 28 October 2015 (UTC)

Document analysis[edit]

The concept of ground truth seems important in digital document analysis. See the work of the Pattern Recognition & Image Analysis Research Lab, University of Salford, Manchester, http://www.primaresearch.org/tools/Aletheia, http://www.primaresearch.org/www/assets/papers/ICPR2010_Pletschacher_PAGE.pdf and elsewhere. This seems to be an important use of the concept, but an entry about this use should be written by somebody of the field (I am not). --Dominique Meeùs (talk) 09:09, 19 November 2015 (UTC)