A1 Vertaisarvioitu alkuperäisartikkeli tieteellisessä lehdessä
Locality kernels for sequential data and their applications to parse ranking
Tekijät: Tsivtsivadze E, Pahikkala T, Boberg J, Salakoski T
Kustantaja: SPRINGER
Julkaisuvuosi: 2009
Journal: Applied Intelligence
Tietokannassa oleva lehden nimi: APPLIED INTELLIGENCE
Lehden akronyymi: APPL INTELL
Vuosikerta: 31
Numero: 1
Aloitussivu: 81
Lopetussivu: 88
Sivujen määrä: 8
ISSN: 0924-669X
DOI: https://doi.org/10.1007/s10489-008-0114-2
Tiivistelmä
We propose a framework for constructing kernels that take advantage of local correlations in sequential data. The kernels designed using the proposed framework measure parse similarities locally, within a small window constructed around each matching feature. Furthermore, we propose to incorporate positional information inside the window and consider different ways to do this. We applied the kernels together with regularized least-squares (RLS) algorithm to the task of dependency parse ranking using the dataset containing parses obtained from a manually annotated biomedical corpus of 1100 sentences. Our experiments show that RLS with kernels incorporating positional information perform better than RLS with the baseline kernel functions. This performance gain is statistically significant.
We propose a framework for constructing kernels that take advantage of local correlations in sequential data. The kernels designed using the proposed framework measure parse similarities locally, within a small window constructed around each matching feature. Furthermore, we propose to incorporate positional information inside the window and consider different ways to do this. We applied the kernels together with regularized least-squares (RLS) algorithm to the task of dependency parse ranking using the dataset containing parses obtained from a manually annotated biomedical corpus of 1100 sentences. Our experiments show that RLS with kernels incorporating positional information perform better than RLS with the baseline kernel functions. This performance gain is statistically significant.
Turun yliopiston Tutkimustietojärjestelmän portaali