Pregled bibliografske jedinice broj: 419323
On Statistical Parameter Setting
On Statistical Parameter Setting // Proceedings of the First Workshop on Psycho-computational Models of Human Language Acquisition, Held in cooperation with COLING-2004 / Sakas, William G. (ur.).
New York (NY): Hunter College, City University of New York, 2004. str. 9-16 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 419323 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
On Statistical Parameter Setting
Autori
Ćavar, Damir ; Herring, Joshua ; Ikuta, Toshikazu ; Rodrigues, Paul ; Schrementi, Giancarlo
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of the First Workshop on Psycho-computational Models of Human Language Acquisition, Held in cooperation with COLING-2004
/ Sakas, William G. - New York (NY) : Hunter College, City University of New York, 2004, 9-16
Skup
Psycho-Computational Models of Human Language Acquisition
Mjesto i datum
Ženeva, Švicarska, 28.08.2009
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
language identification; algorithm; statistics; n-gram model; unsupervised learning
Sažetak
We present a model and an experimental platform of a bootstrapping approach to statistical induction of natural language properties that is constraint based with voting components. The system is incremental and unsupervised. In the following discussion we focus on the components for morphological induction. We show that the much harder problem of incremental unsupervised morphological induction can outperform comparable all-at-once algorithms with respect to precision. We discuss how we use such systems to identify cues for induction in a cross-level architecture.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo, Informacijske i komunikacijske znanosti, Filologija
