Seminar on Computational Learning and Adaptation


  Memory-based Learning for Article Generation

Francis Bond
NTT Machine Translation Research Group
Kyoto, Japan
www.kecl.ntt.co.jp/icl/mtg/

Article choice can pose difficult problems in applications such as machine translation and automated summarization. In this paper, we investigate the use of corpus data to collect statistical generalizations about article use in English in order to generate articles automatically to supplement a symbolic generator. We use data from the Penn Treebank as input to a memory-based learner (TiMBL 3.0; Daelemans et al., 2000), which predicts whether to generate an article with respect to an English base noun phrase. We discuss competitive results obtained using a variety of lexical, syntactic and semantic features that play an important role in automated article generation.

Minnen, Guido, Francis Bond, and Ann Copestake (2000). Memory-based Learning for Article Generation. In Proceedings of the Fourth Conference on Computational Natural Language Learning and of the Second Learning Language in Logic Workshop (CoNLL-2000 and LLL-2000), Lisbon. 43-48.


Date: Thurs., Oct 26

Time: 4:15-5:30PM

Place: Cordura 100


Return to the seminar schedule