@article{4616,
  author = {Chen Wang},
  title = {Modeling Clustering Approaches to Recommender Systems in Language Datasets},
  journal = {Journal of Data Processing},
  year = {2025},
  volume = {15},
  number = {4},
  doi = {https://doi.org/10.6025/jdp/2025/15/4/175-182},
  url = {https://www.dline.info/jdp/fulltext/v15n4/jdpv15n4_3.pdf},
  abstract = {This paper proposes a novel approach to improving personalized recommendations in English resource
libraries. Addressing challenges such as information overload, lack of personalization, cold start problems,
and algorithmic complexity, the study leverages the K-means clustering algorithman unsupervised machine
learning technique to group users and resources based on similarity. By transforming multidimensional resource
attributes (e.g., topic, proficiency level, target audience) into onedimensional data via dimensionality
reduction, the system improves storage, search, and recommendation efficiency. The model integrates user
English proficiency data, processes it via dynamic multimodal modeling and principal component analysis,
and clusters users for tailored suggestions. Experimental results demonstrate that the K-means based system
outperforms traditional collaborative filtering in recommendation accuracy, although recall rates vary
with list size. User activity data reveal consistent personalized recommendation usage (~40% daily peak)
and search spikes at 15:00 and 21:00, likely reflecting student and working user learning patterns. The study
concludes that the proposed method effectively enhances resource management and user satisfaction, ofering
a scalable and efficient solution for digital English libraries. Future work includes refining algorithms
to improve accuracy and user experience further.},
}