: NLP Vocabulary: Spearman’s rank correlation

NLP resources and evaluation > measure > Spearman’s rank correlation

Spearman’s rank correlation

Spearman's rank correlation coefficient denoted by r is a non-parametric metric to measure statistical dependence between ranking of two variables. (Chatterjee, Maheshwari, Ramakrishnan & Jagarlapudi, 2021)
The Spearman correlation coefficient is a nonparametric measure of rank correlation assessing how well the relationship between two variables can be described. (Yin, Wang, Li, Li & Wang, 2016)

Evaluation is done by comparing the new created ratings against the test (gold) ratings using Spearman's rank-order correlation. (Köper & Schulte im Walde, 2017)
For comparing our model's predictions with the scores assigned by human subjects we use the Spearman correlation coefficient (Spearman 1904) a well-established nonparametric measure of rank correlation between two variables. (Haider, 2018)
For question formation the Spearman rank-order correlation between the number of parameters and accuracy is 0.51 (sequence) and 0.58 (main auxiliary); for passivization 0.75 (sequence) and 0.43 (object). (Mueller & Linzen, 2023)
In order to get an idea of how closely the respective metrics are related the Spearman rank correlation coefficients were calculated for all automatic evaluation metric combinations. (Paul, 2008)
It turns out that the average Spearman rank-order correlation is not sensitive enough to errors for the purposes of comparing favourable disagreement within ensembles. (Curran, 2002)

http://data.loterre.fr/ark:/67375/8LP-TP0T6RZQ-S

RDF/XML TURTLE JSON-LD Last modified 5/29/24

Loterre