Leksika-grammatika piirimail: publitsistlike arvamuslugude keelekasutusmustrid
At the border of lexis and grammar: Language use patterns of journalistic opinion articles

Summary/Abstract: The study draws on the usage-based corpus linguistic approaches. It presents the linguistic analysis of opinion articles (published in the Estonian daily newspapers Postimees and Õhtuleht), relying on part-of-speech trigrams. These language use patterns have been extracted from morpho-syntactically tagged texts using the data mining software Cluster Catcher (Klastrileidja) developed at the Tallinn University. Earlier, similar analysis has been done based on the corpora of Estonian fiction and learner language. The recurring part-of-speech sequences have the encoded phonological, syntactic, and conceptual rules of their components’ co-occurrence that manifest at the border of semantics and grammar, bringing out the regularities of language use. Therefore, the actively used language patterns can be considered a suitable source material for teaching and learning Estonian, comparing the language production of learners speaking different native languages and also for automated processing of learner writings (e.g., proposing corrections in wording with references to rules and examples of the target language use).

• Details
• Contents