KazCausal: Manual Annotation of Causal Constructions in the Kazakh Language

Roman Taberkhan, Madina Sambetbayeva

2025

Closed access Metadata (CSL-JSON)
DOI 10.1109/ubmk67458.2025.11206832
Data confidence 85 high
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

This article presents KazCausal, a manual annotation scheme developed to identify causal constructions in Kazakh language texts. The research aims to create a formalized corpus that reflects the morphological, syntactic, and semantic aspects of causality expression in the agglutinative structure of the Kazakh language. The scheme supports multiple labels and overlapping relationships, such as direct consequence, motivation, sanction, emotional conditioning, and indefinite causality. The corpus includes 681 sentences, 150 of which are marked as containing causal relationships. Inter-annotator agreement analysis demonstrates a high level of consistency. The results can be used in NLP tasks, including model training and automatic extraction of causal relationships.