lausekogumid

Lausekogumid are curated collections of sentences used in Estonian linguistics. They are assembled to illustrate language use, grammar, and meaning. They can be monolingual, containing Estonian sentences, or bilingual, containing translations. They are employed in language teaching, linguistic research, and natural language processing.

The main aims are to demonstrate grammatical phenomena (such as case, agreement, tense), to provide authentic usage examples, and to support learners through varied registers from everyday speech to formal writing.

Contents and formats vary. A lausekogum may group sentences by structure or topic and attach metadata, including source, author, date, intended audience, and difficulty. Files are usually plain text but may be encoded in JSON, XML, or specialized annotation formats.

Curation involves criteria like representativeness, coverage of constructions, correctness, and inclusivity. Annotations (where present) may include part-of-speech tagging, morphological analysis, or syntactic trees. Licensing ranges from closed to open licenses, with many educators preferring open-licensed collections.

In related resources, lausekorpus refers to large corpora of sentences collected from real texts, while näidislaused are individual model sentences used as examples. Lausekogumid thus complement corpora and dictionaries by providing ready-made exemplars for study and practice.