TY - JOUR
AU - 
AB - Texts Candidate Extraction We propose an event-driven model for Phrases Events Sentences headline generation. Given an input document, the system identifies a key event chain by extracting a set of structural Candidate Ranking events that describe them. Then a novel multi-sentence compression algorithm ... ... is used to fuse the extracted events, Candidate #1 Candidate #i Candidate #K generating a headline for the document. Our model can be viewed as a novel Multi-Sentence Compression combination of extractive and abstractive headline generation, combining the advantages of both methods using event Headline Headline Generation structures. Standard evaluation shows that our model achieves the best performance Figure 1: System framework. compared with previous state-of-the-art systems. Zajic et al., 2005). Abstractive models choose a 1 Introduction set of informative phrases for candidate extraction, and then exploit sentence synthesis techniques for Headline generation (HG) is a text summarization headline generation (Soricut and Marcu, 2007; task, which aims to describe an article (or a set of related paragraphs) using a single short sentence. Woodsend et al., 2010; Xu et al., 2010). The task is useful in a number of practical Extractive HG and abstractive HG have scenarios, such as compressing text for mobile their respective 
TI - Event-Driven Headline Generation
JF - Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
DO - 10.3115/v1/p15-1045
DA - 2015-01-01
UR - https://www.deepdyve.com/lp/unpaywall/event-driven-headline-generation-r00IEBkIDq
DP - DeepDyve
ER -