Multi-Level Text Clustering in Subject Knowledge Library and its Visualization

Abstract

The large-scale and complex data generated in the teaching field of business administration poses challenges for decision-makers and managers of companies, and how to effectively extract and manage the useful information contained in these data has become a problem to be solved. Currently available methods of subject knowledge library clustering and visualization struggle to handle the complexity and multi-hierarchies of such subject data effectively or meet users’ requirements for advanced semantic understanding and retrieval. In view of these matters, this study aims to probe deeper into the problem of multi-level text clustering in the subject knowledge library and its visualization. Firstly, an innovative strategy-based subject semantic representation method for knowledge libraries was proposed to better interpret and represent the semantic information of subject data. Secondly, a subject clustering model of the knowledge library was constructed based on an improved hierarchical Dirichlet polynomial distribution, enabling efficient and accurate clustering of subject data. Lastly, visualization technology was employed to display the cluster results, allowing users to gain a clear understanding of the internal relationships and structure of the subject data. The research findings of this study could provide valuable new tools and methods for solving the problem of subject knowledge library management and utilization, analyzing the subject data, and supporting decision-making. As a result, they hold both theoretical and practical significance.