Using machine learning to generate headlines

被引：0

作者：

Wang, RC ^{[1
]}

Stokes, N ^{[1
]}

Doran, W ^{[1
]}

Dunnion, J ^{[1
]}

Carthy, J ^{[1
]}

机构：

[1] Univ Coll Dublin, Dept Comparat Biosci, Intelligent Informat Retrieval Grp, Dublin 2, Ireland

来源：

MLMTA '05: Proceedings of the International Conference on Machine Learning Models Technologies and Applications | 2005年

关键词：

machine learning; headline generation; statistical techniques; DUC;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

In this paper, we present a system which generates headlines for a text. Our system, the HybridTrim system, uses a linguistic, statistical and positional information in combination with a machine learning technique to identify topic labels for headlines in a text. In this paper, we compare our system with the Topiary system which, in contrast, uses a statistical learning approach to finding topic descriptors for headlines. Both 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.

引用

页码：167 / 172

页数：6

共 9 条

[1]

[Anonymous], 1998, C5. 0: An informal tutorial

[2]

DORAN W, 2004, P CICLING SEOUL

[3]

Doran W. P., 2004, Proceedings of Sheffield SIGIR 2004. The Twenty-Seventh Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, P464, DOI 10.1145/1008992.1009072

[4]

DORAN WP, 2004, P DOC UND C DUC

[5]

Dorr Bonnie, 2003, P HLT NAACL

[6]

LIN CY, 2004, P ACL WORKSH TEXT SU, P56

[7]

LIN CY, 2003, P HLT NACCL

[8]

Stokes Nicola, 2004, Ph. D. Dissertation.

[9]

Zajic D., 2004, P DOC UND C DUC

← 1 →