Clustering analysis of Russian satirical literary texts based on an improved K-means algorithm

This study presents a clustering analysis of Russian satirical literature texts using an improved K-means algorithm. Russian satire, rich in linguistic complexity and subtlety, offers unique challenges for text classification and clustering. The proposed algorithm enhances the traditional K-means method by integrating a weighted term frequency mechanism and optimizing the initialization process to better capture the nuances of satirical language. The clustering process is conducted using a corpus of Russian satirical texts, with the aim of categorizing them into meaningful clusters based on themes and stylistic features. Experimental results demonstrate that the improved K-means algorithm outperforms conventional clustering techniques in terms of accuracy and relevance, highlighting its potential for text mining in literary analysis. This approach provides a novel tool for researchers and practitioners in the fields of computational linguistics, literary studies, and cultural analysis.