Application of an intelligent English text classification model with improved KNN algorithm in the context of big data in libraries

作     者：Xu, Qinwen 

作者机构：Henan Med Coll Publ Fdn Dept Zhengzhou 451191 Peoples R China 

出 版 物：《SYSTEMS AND SOFT COMPUTING》 (Syst. Soft. Comput.)

年 卷 期：2025年第7卷

核心收录：

主　　题：K -means clustering KNN algorithm Category average distance Intelligent classification of English texts Big data analysis 

摘      要：In the era of big data, libraries manage huge electronic text resources, of which English text resources are particularly critical for academic research, student learning, and professional knowledge acquisition. This paper aims to improve the K-nearest neighbor algorithm and design an intelligent classification model to improve the efficiency and quality of library services. An improved method based on in-class K-means clustering and class mean distance is used to characterize and extract text information with a vector space model. The results showed that the improved K-nearest neighbor algorithm achieved significant improvement in the precision, recall, and F1 values, reaching 90.50 %, 89.95 %, and 89.37 %, respectively. The classification time was significantly reduced to 1034.57 s. In addition, the improved algorithm had a classification accuracy of 94 %, surpassing other popular text classification algorithms. The research successfully realizes the efficient classification of text. The research results not only improve the classification efficiency of library English text resources but also provide strong support for readers to quickly obtain the required information, which has important application value and wide application prospects.