A Bayesian Approach to Query Language Identification

Warning

This publication doesn't include Faculty of Education. It includes Faculty of Informatics. Official publication website can be found on muni.cz.

Authors	MATERNA Jiří HREŠKO Juraj
Year of publication	2011
Type	Article in Proceedings
Conference	Recent Advances in Slavonic Natural Language Processing
MU Faculty or unit	Faculty of Informatics
Citation
Web	https://nlp.fi.muni.cz/raslan/2011/paper10.pdf
Field	Informatics
Keywords	language identification; query language; information retrieval
Description	In this paper we present a Bayesian approach to language identification of queries sent to an information retrieval system. The aim of the work is to identify both the language of a query as a whole and the language of particular words in the query. The method is evaluated on a test set of manually labelled queries. The evaluation shows that our method performs better than the Google Language Detect API and an implementation of the n-gram method on our testing set of queries.
Related projects:	Centrum komputační lingvistiky