hrcak mascot   Srce   HID

Izvorni znanstveni članak

25 years of Hašek

Šandor Dembitz

Puni tekst: hrvatski, pdf (829 KB) str. 138-150 preuzimanja: 20* citiraj
APA 6th Edition
Dembitz, Š. (2019). 25 godina Hašeka. Jezik, 66 (4-5), 138-150. Preuzeto s
MLA 8th Edition
Dembitz, Šandor. "25 godina Hašeka." Jezik, vol. 66, br. 4-5, 2019, str. 138-150. Citirano 12.07.2020.
Chicago 17th Edition
Dembitz, Šandor. "25 godina Hašeka." Jezik 66, br. 4-5 (2019): 138-150.
Dembitz, Š. (2019). '25 godina Hašeka', Jezik, 66(4-5), str. 138-150. Preuzeto s: (Datum pristupa: 12.07.2020.)
Dembitz Š. 25 godina Hašeka. Jezik [Internet]. 2019 [pristupljeno 12.07.2020.];66(4-5):138-150. Dostupno na:
Š. Dembitz, "25 godina Hašeka", Jezik, vol.66, br. 4-5, str. 138-150, 2019. [Online]. Dostupno na: [Citirano: 12.07.2020.]

Hašek is a Croatian on-line spellchecker that continuously operates since March 21, 1994,
nowadays at the address In 25 years of functioning Hašek processed
nearly 30 million texts, which build a corpus of more than 7 billion tokens. By comparison,
all books ever published in Croatian form a corpus with less than 20 billion tokens.
As a WWW-embedded tool, Hašek took advantage of many web-based services including
learning. Thanks to Hašek’s learning capability, its dictionary increased from initial 100
thousand to more than 2 million word-types. Another aspect of learning was the creating
and regular updating of the Croatian n-gram system. Unlike Google, whose n-gram systems
are based on the WaC (Web as Corpus) approach and cut-off criteria, Croatian n-grams
were extracted from processed texts by a lexical criterion: each n-gram constituent must
be proven by the spellchecker as valid in Croatian spelling. The difference in approaches
made Croatian n-gram system comparable in size to the largest Google n-gram systems.
Unfortunately, the advantages of on-line spellchecking for rapid breakthroughs into much
more sophisticated language technology areas were not recognized by Croatian decision
makers, with some consequences mentioned in the paper.

Ključne riječi
Hašek; spellchecking; learning; Google; n-gram systems

Hrčak ID: 237727



Posjeta: 50 *