Fig. 3

ROUGE evaluation scores for different extractive summarization methods using abstracts as gold standard summaries. a Performance comparison of different summarization approaches on CRAFTtest corpus (32 full-text scientific articles); b Performance comparison of different summarization approaches using ROUGE evaluation toolkit on clinical case reports corpus. The p-values for ROUGE-1, ROUGE-2, and ROUGE-SU4 scores for each method were compared against respective baseline scores using a one-sided Wilcox text. The MINTS algorithm outperformed other methods in both cases