: NLP Vocabulary: false negative rate

NLP resources and evaluation > measure > confusion matrix > false negative rate

false negative rate

The proportion of positives which yield negative test outcomes with the test, i.e., the conditional probability of a negative test result given that the condition being looked for is present. (Wikipedia)

Besides the focus on Dutch text it is desirable for high-impact applications like those used in medical practice and law enforcement to work with models and algorithms that have low false negative rates due to potential societal and ethical complications that arise with false positives. (Loerakker, Müter & Schraagen, 2024)
Hate speech aimed at particular ideologies such as partisan intellectual or sports affiliations had a high false negative rate (FNR) for both the English and Korean hate speech classifiers. (Lee, Jung & Oh, 2023)
In addition they produce considerably higher false negative rates with a magnitude up to five times greater demonstrating the extent of the cultural gap. (Lee, Jung & Oh, 2023)
Investing further effort in lexicon creation may have further reduced the false negative rate. (Bialer, Izmaylov, Segal, Tsur, Levi-Belz & Gal, 2022)
Similarly we calculate the false negative rate for samples where the LMs prefer antistereotypical keywords for the original group but prefer it even more for the control group. (Pikuliak, Beňová & Bachratý, 2023)

http://data.loterre.fr/ark:/67375/8LP-SS7J8HK6-4

RDF/XML TURTLE JSON-LD Last modified 6/26/24

Loterre