Abstract
This paper presents a study of named entity recognition (NER) in the Kazakh language using the transformer model of the spaCy library. There are a few works have been done on the NER in the Kazakh language and there is only one open data set. This indicates that the Kazakh language is a low-resource language and the relevance of the study. The article considered all scientific papers written so far on the NER in the Kazakh language and defined the main problems of subject area based on analyzing of recently published research. It provides a formal description of the transformer model from spaCy for NER. To identify the named entities from Kazakh-language texts, a data set was prepared from scientific works, and the recognition of named entities were carried out in a complete pipeline sequence. The main focus was on preparing annotated data for implementing NER from Kazakh texts and on the pipeline NER component.