Application of K-Means Clustering Algorithm in Automatic Machine Learning

This article studies the application of k-means clustering algorithm in automatic machine learning. By selecting the Iris dataset from the UCI dataset for experiments, we found that the k-means algorithm can accurately classify data. At the same time, we compared the traditional k-means clustering algorithm with the optimized K-Means++ and Spherical K-Means clustering algorithms in terms of clustering accuracy and speed, and found that the optimized algorithm has improved clustering accuracy and speed compared to traditional algorithms. Therefore, applying these two optimized k-means clustering algorithms in AutoML can improve the clustering and generalization abilities of machine learning models.