This paper presents a comparative study on two key problems existing in extrac-tive summarization: the ranking problem and the selection problem. To this end, we presented a systematic study of comparing different learning-to-rank al-gorithms and comparing different selec-tion strategies. This is the first work of providing systematic analysis on these problems. Experimental results on two benchmark datasets demonstrate three findings: (1) pairwise and listwise learn-ing- to-rank algorithms outperform the baselines significantly; (2) there is no significant difference among the learn-ing- to-rank algorithms; and (3) the in-teger linear programming selection strategy generally outperformed Maxi-mum Marginal Relevance and Diversity Penalty strategies.
展开▼
