Abstract
In this paper, we present the useful features of a syntactic constituent for a probabilistic parsing model and analyze the combination of the features in order to disambiguate parse trees effectively. Unlike most of previous works focusing on the features of a single head, the features of a functional head, the features of a content head, and the features of size are utilized in this study. Experimental results show that the combination of different features such as the functional head feature and the size feature is prefered to the combination of similar features such as the functional head feature and the content head feature. Besides, it is remarkable that the function feature is more useful than the combination of the content feature and the size feature.
| Original language | English |
|---|---|
| Pages (from-to) | 97-101 |
| Number of pages | 5 |
| Journal | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
| Volume | 2945 |
| Publication status | Published - 2004 Dec 1 |
Fingerprint
ASJC Scopus subject areas
- Computer Science(all)
- Biochemistry, Genetics and Molecular Biology(all)
- Theoretical Computer Science
Cite this
Evaluation of feature combination for effective structural disambiguation. / Park, So Young; Kwak, Yong Jae; Lim, Joon Ho; Rim, Hae-Chang.
In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 2945, 01.12.2004, p. 97-101.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Evaluation of feature combination for effective structural disambiguation
AU - Park, So Young
AU - Kwak, Yong Jae
AU - Lim, Joon Ho
AU - Rim, Hae-Chang
PY - 2004/12/1
Y1 - 2004/12/1
N2 - In this paper, we present the useful features of a syntactic constituent for a probabilistic parsing model and analyze the combination of the features in order to disambiguate parse trees effectively. Unlike most of previous works focusing on the features of a single head, the features of a functional head, the features of a content head, and the features of size are utilized in this study. Experimental results show that the combination of different features such as the functional head feature and the size feature is prefered to the combination of similar features such as the functional head feature and the content head feature. Besides, it is remarkable that the function feature is more useful than the combination of the content feature and the size feature.
AB - In this paper, we present the useful features of a syntactic constituent for a probabilistic parsing model and analyze the combination of the features in order to disambiguate parse trees effectively. Unlike most of previous works focusing on the features of a single head, the features of a functional head, the features of a content head, and the features of size are utilized in this study. Experimental results show that the combination of different features such as the functional head feature and the size feature is prefered to the combination of similar features such as the functional head feature and the content head feature. Besides, it is remarkable that the function feature is more useful than the combination of the content feature and the size feature.
UR - http://www.scopus.com/inward/record.url?scp=35048823897&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=35048823897&partnerID=8YFLogxK
M3 - Article
VL - 2945
SP - 97
EP - 101
JO - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
JF - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SN - 0302-9743
ER -