Lingsoft's Grammar Checker Module for Norwegian (Bokmål)
|Lingsoft® NOBGRC is designed for checking basic grammar and style errors in standard written Norwegian (Bokmål). Lingsoft pioneered in applying constraint grammar for designing grammar checker rules, that are aware of word and sentence structures of the corresponding language.|
NOBGRC detects potential grammar and writing convention errors as well as stylistically non-recommended expressions from more than 40 error categories, presupposing that the context where the error or the non-recommended expression occurs conforms to the context condition of the error detection rule. The context conditions may be based on specific sequences of words, inflected forms or individual characters.
Most error-detection rules will attempt to provide one or more potential corrections for the detected error. In some cases correcting an error may cause a further error, typically an agreement error, to be detected. With some types of errors, suggesting a correction is not feasible. In these cases the grammar checker delivers an explanation of the detected error.
In the design and development of NOBGRC, Lingsoft has endeavored to minimize the number of linguistic errors, so that an adequate balance is reached between error detection and false alarm avoidance, as far as such a balance can be reached by computational means. In particular, Lingsoft has endeavored to achieve the following gold-standard metric:
The number of false flags detected by the grammar checker will be less than one (1) per standard A4/letter page (consisting of single-spaced, unformatted 12-point font-size text).
It all starts from analyzing the wordsNOBGRC uses Lingsoft's comprehensive two-level model of Norwegian (Bokmål) morphology for assigning grammatical tags to the analyzed words, and for generating particular inflected forms to error suggestions. The model contains more than 110,000 lexical entries, covering the central vocabulary of Norwegian (Bokmål), including abbreviations, acronyms, proper names and numerals. Two-level rules govern word transformation issues like book, bøker (book, books).
For more reliable detection of grammar errors, NOBGRC input is disambiguated with Lingsoft's constraint grammar parser for Norwegian (Bokmål) after being analyzed with NOBTWOL.
The lexical content and two-level rules of the language model are compiled to a fast and compact finite-state transducer, which together with the disambiguation and error-detection rules and other program code and data are included in a binary file of only about 3 MB.
Software integration made easyNOBGRC can be integrated to provide grammar and style-checking to almost any software application, including web-based services, with Lingsoft's proprietary LSPROOF-API application programming interface for Windows, Linux, Mac and Java. The character set used with LSPROOF is Unicode.
How to buy
NOBGRC: Copyright © Lingsoft, Inc. 2010. Two-Level Compiler: Copyright © Xerox Corporation 1994. Lingsoft is a registered trademark and NOBGRC, NOBTWOL and LSPROOF are trademarks of Lingsoft, Inc. All rights reserved. Details subject to change.
Copyright ©1986-2018, Lingsoft Ltd.