The Application of Cluster Analysis Algorithms for Student Assessment Data in Educational Quality Management

Abstract

The implementation of teaching quality management plays a guiding role in a series of educational tasks, including teaching reform; selecting appropriate evaluation methods is crucial for the effective implementation of teaching quality management. This paper combines student grade data with an ant colony clustering algorithm, which establishes a “global memory bank of historical positions” to guide ants in purposefully and rapidly depositing the data objects they carry, while also effectively avoiding the occurrence of local optima. Finally, the ant colony clustering algorithm was applied to student evaluation and experimentally compared with the K-Means algorithm. The experimental results show that, compared to K-Means, the ant colony clustering algorithm improved the contour coefficient by 0.1009 (48.53%) and reduced the DBI index by 0.6823 (32.84%) in the clustering analysis of student evaluation data, significantly enhancing stability. Three-dimensional visualization further validated the clustering structure characterized by intra-cluster compactness and distinct inter-cluster separation. This method provides data support for precise student evaluation and differentiated instruction, demonstrating promising application prospects.