Analysis of emotional coloring of Karakalpak news articles in local online media

Abstract

This paper presents an original approach to sentiment analysis of news articles in the Karakalpak language using a hybrid neural architecture that combines layers of convolutional neural networks and bidirectional LSTMs. The Karakalpak language, being one of the poorly studied Turkic languages, is of considerable interest to researchers, in particular due to the low resource content of this language. To implement the task, a corpus of 10,000 sentences from local news sources was collected, each of which was labeled as positive, negative or neutral. At the same time, the model was trained for 18 epochs, however, using the early stopping function, the weights were rolled back to the results of the 13th epoch. The model was tested on three different test sets, where the results ranged from 75% to 83% according to the f1-score metric.

More by Nilufar Abdurakhmonova

• A model of a universal linguistic knowledge graph for Turkic languagesArctic XXI century · 2026
• Machine Learning-Based Topic Classification of Economic and Financial Texts in UzbekInternational Conference on Computer Science and Engineering · 2026
• Building a Phraseological Ontology of Uzbek Idioms and Evaluating Similarity Measures for Their Detection in TextInternational Conference on Computer Science and Engineering · 2026
• Developing Morphological and Syntactic Analysis Models for Karakalpak as a Low-Resource Turkic LanguageInternational Conference on Computer Science and Engineering · 2026