ELAN as a search engine for hierarchically structured, tagged corpora
Joshua Wilbur · 2019
The main goal of this paper is to outline and explore the usefulness of the corpus search functionalities provided in the ELAN annotation application when annotations are provided in hierarchically organized tiers.A general overview of ELAN's search functions is provided first, highlighting the program's usefulness as a corpus search engine for corpus and computational linguists.To illustrate this, the updated hierarchical tier structure for ELAN developed by the Freiburg Research Group in Saami Studies for the group's projects on both Saamic and Komi languages is presented as an example template.The suitability of hierarchical structures for annotations and the ELAN search interfaces for doing corpus linguistics is explored critically, including the description of a fundamental flaw in the "Multiple Layer Search" mode which likely prevents ELAN from being used as a search engine for complex corpus studies. KokkuvõteArtikli peamine eesmärk on kirjeldada ning uurida, millised on programmi ELAN korpusepäringu võimalused, kui materjal on annoteeritud hierarhiliste kihtidena.Selleks antakse kõigepealt ülevaade programmi otsimootori üldistest võimalustest, tuues välja selle kasulikud omadused korpus-ja arvutilingvistide jaoks.Näitena tutvustatakse ELAN-i uuendatud hierarhilist kihistruktuuri, mille on välja arendanud Freiburg Research Group in Saami Studies töötades nii saami keelte kui ka komi keele teadusprojektidega.Artiklis arutletakse selle üle, kuivõrd