TY - CONF ID - ZopfLozaFuernkranz2016CPSum T1 - Beyond Centrality and Structural Features: Learning Information Importance for Text Summarization A1 - Zopf, Markus A1 - Loza Mencía, Eneldo A1 - Fürnkranz, Johannes TI - Proceedings of the 20th SIGNLL Conference on Computational Natural Language Learning Y1 - 2016 SP - 84 EP - 94 PB - Association for Computational Linguistics CY - Berlin, Germany UR - http://www.aclweb.org/anthology/K16-1009 N2 - Most automatic text summarization systems proposed to date rely on centrality and structural features as indicators for information importance. In this paper, we argue that these features cannot reliably detect important information in heterogeneous document collections. Instead, we propose CPSum, a summarizer that learns the importance of information objects from a background source. Our hypothesis is tested on a multi-document corpus where we remove centrality and structural features. CPSum proves to be able to perform well in this challenging scenario whereas reference systems fail. ER -