Abstract
AbstractThe Kmeans clustering algorithm is widely used for the advantages of simplicity and efficient operation. However, the lack of clustering centers in the algorithm usually causes incorrect category of some discrete points. Therefore, in order to obtain more accurate clustering results when studying the factors affecting the professional growth of outstanding teachers, this paper proposes an improved algorithm of Kmeans combined with DBSCAN. Observing the clustering results of the influencing factors and calculating the evaluation standard values of the clustering results, it is found that the optimized DB-Kmeans algorithm has obvious improvements in the accuracy of the clustering results, and the clustering effect of the algorithm on edge points is more advantageous than the original algorithms according to the scatter diagram.
Funder
Scientific Research Project of Tianjin Educational Committee
Publisher
Springer Science and Business Media LLC
Reference11 articles.
1. Y. Wu, Research on Passenger Car Passenger Order Scheduling Based on Canopy-Kmeans Algorithm (Shanxi University, 2020)
2. X. Gao, An Improved K-means Clustering Algorithm and a New Clustering Effectiveness Index Research (Anhui University, 2020)
3. M. Yan, X. Xie, W. Li, D. Wu, X. Cui, S. Pan, Morphological clustering algorithm of typical load curve based on Gaussian kernel density estimation. Electr. Meas. Instrum. 1–8 (2022). https://kns.cnki.net/kcms/detail/detail.aspx?dbcode=CAPJ%26dbname=CAPJLAST%26filename=DCYQ20210316003%26uniplatform=NZKPT%26v=fzYhayA03xb9KgYHcfE22jsZ7B3eNVIqtoqr0ToI60YAoWnfgwuDsWQj7-MOLMZ
4. G. HimaBindu, Ch. Raghu Kumar, C. H. Hemanand, N. Rama Krishna, Hybrid clustering algorithm to process big data using firefly optimization mechanism. Mater. Today Proc. (2020). https://doi.org/10.1016/j.matpr.2020.10.273
5. N. Valarmathy, S. Krishnaveni, A novel method to enhance the performance evaluation of DBSCAN clustering algorithm using different distinguished metrics. Mater. Today Proc. (2020). https://doi.org/10.1016/j.matpr.2020.09.623
Cited by
3 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献