ROUGE-N Score

#nlp #deep-learning #interview

ROGUE = Recall-Oriented Understudy for Gisting Evaluation
As ROGUE compare with the ALL target sentences, it is often compared with Recall
ROGUE-N score is the sum of Recall of N-grams
Heavily used in Text Summarization, Also usually used in Machine Translation with BLEU Score

ROGUE Score

$ROGUE-N = \sum_{i = 1}^{N} R e c a l l_{i}$ ${Recall}_{n} = \frac{# of n-grams matched both on generation and target}{# of n-grams in target}$

Problems with ROGUE Score

Doesn't consider semantic meaning
Hard to compare with different tokenizers
Doesn't consider synonyms

Related Notes