: NLP Vocabulary: mean absolute error

NLP resources and evaluation > measure > mean absolute error

mean absolute error

Mean Absolute Error (MAE) was also computed for the systems with respect to manual labels. (Lucas, Debnath & Hines, 2024)
The aggregated Mean Absolute Error (MAE) is 8.5 when using all the features. (Soricut & Echihabi, 2010)
The MAE is calculated over 3 random splits of the data with 809 812 and 811 documents in the test set respectively. (Sim, Routledge & Smith, 2015)
The MAE scores for the level labels are all far below 1 both for the 0-4 scales and the 0-5 scales. (Kim, Verkijk, Geleijn, van der Leeden, Meskers, Meskers, van der Veen, Vossen & Widdershoven, 2022)

http://data.loterre.fr/ark:/67375/8LP-RCQ5DPNL-1

RDF/XML TURTLE JSON-LD Last modified 6/3/24

Loterre