How to "selecting the best subset of sentences using a greedy approach"

[perthrocky]

Hi Shashi:

I would like to train your model on another dataset.
In preprocessing CNN datasets, every article has 2 or 3 sentences with ROUGE score.
How do you select these 2 or 3 sentences?
In your paper, you just mention the greedy approach with ROUGE score, would you
kindly give me more hints or details for this greedy approach?

Thanks

[shashiongithub]

Check section 4.2: We approximate Yˆ by the k(=5) extracts which receive highest ROUGE scores.
More concretely, we assemble candidate summaries efficiently by first selecting p(=10) sentences
from the document which on their own have high ROUGE scores. We then generate all possible
combinations of p sentences subject to maximum length m and evaluate them against the gold summary. Summaries are ranked according to F1 by taking the mean of ROUGE-1, ROUGE-2, and
ROUGE-L.

Let me know if it not clear.