@article{4581,
  author = {Yinhui Hao, Chunyan Zhao},
  title = {Optimizing the Student's Exercise Behavior Analysis: A Clustering Model},
  journal = {Journal of Information Technology Review},
  year = {2025},
  volume = {16},
  number = {4},
  doi = {https://doi.org/10.6025/jitr/2025/16/4/129-137},
  url = {https://www.dline.info/jitr/fulltext/v16n4/jitrv16n4_1.pdf},
  abstract = {This paper proposes an optimized Ant Colony Algorithm (ACA) model to cluster and mine the physical exercise behavior characteristics of college students. Recognizing that students’ unstructured exercise habits often lead to inefficacy or injury, the study aims to provide more accurate behavioral analysis. The model first constructs a behavior representation by tracking the centroid movement of students’ bodies in video sequences to generate feature vectors. It then employs an ACA, inspired by ants’ foraging behavior, to cluster these vectors efficiently. The algorithm is optimized to address common ACA issues like slow convergence and stagnation by improving pheromone updating and path selection. Experimental results on standard and custom datasets show the model achieves higher F-measure values, greater stability, and faster convergence compared to other algorithms, effectively avoiding local optima for superior clustering accuracy.},
}