Table 1 ROUGE scores measure the summarization quality.

From: Automatic summarization model based on clustering algorithm

Model

ROUGE-1

ROUGE-2

ROUGE-l

LEAD-3

40.34

17.70

36.57

JECS

41.70

18.50

37.90

LSTM + PN

41.85

18.93

38.13

Longformer-Ext

43.00

20.20

39.30

BERTSUMEXT

43.25

20.24

39.63

MatchSum

42.82

18.11

38.79

ClusterSum

44.22

20.26

40.12

  1. ROUGE-1, ROUGE-2 and ROUE-l are used for reporting the unigram, bigram and longest common subsequence overlap with reference summaries.
Back to article page