ROUGE-L Score

#deep-learning #nlp #interview

ROGUE = Recall-Oriented Understudy for Gisting Evaluation
As ROGUE compare with the ALL target sentences, it is often compared with Recall
Better at comparing semantic meaning than ROUGE-N Score
Heavily used in Text Summarization, Also usually used in Machine Translation with BLEU Score

ROGUE-L Score

$ROGUE-L = \frac{Length of ALL LCS (same size) on both generation and target}{# of words in target}$

Problems with ROGUE Score

Hard to compare with different tokenizers
Doesn't consider synonyms

Related Notes