A hybrid statistical/linguistic approach to headline generation

Author

Wang, Rui-Chao ; Stokes, Nicola ; Doran, William ; Dunnion, John ; Carthy, Joe

Author_Institution

Dept. of Comput. Sci., Univ. Coll. Dublin, Ireland

Volume

4

fYear

2005

fDate

18-21 Aug. 2005

Firstpage

1971

Abstract

In this paper, we present the HybridTrim system, which uses a machine learning technique to combine linguistic, statistical and positional information to identify topic labels for headlines in a text. We compare our system with the Topiary system which, in contrast, uses a statistical learning approach to finding topic descriptors for headlines. Both of these systems combine these topic descriptors with a compressed version of the lead sentence. The performance of these systems is evaluated using the ROUGE evaluation suite on the DUC 2004 news stories collection.

Keywords

abstracting; computational linguistics; learning (artificial intelligence); text analysis; HybridTrim system; ROUGE evaluation suite; Topiary system; computational linguistics; headline generation; hybrid statistical-linguistic approach; machine learning technique; statistical learning approach; topic label identification; Computational linguistics; Computer science; Educational institutions; Electronic mail; Hybrid power systems; Information retrieval; Learning systems; Machine learning; Statistical learning; Statistics; Headline generation; computational linguistics; machine learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on

Conference_Location

Guangzhou, China

Print_ISBN

0-7803-9091-1

Type

conf

DOI

10.1109/ICMLC.2005.1527268

Filename

1527268