Regional enterprise economic development dimensions based on k-means cluster analysis and nearest neighbor discriminant

Abstract

In the analysis of the dimension of economic development, the clustering method has a strong dependence on the selection of the central point. However, the random selection of the central point by the traditional K-means clustering method involves the most sensitive central point selection problem of the clustering method. In order to improve the economic development dimension analysis, this paper improves the initial center selection method based on the traditional K-means clustering method, so that the traditional K-means clustering method is no longer a random selection of initial center, and the problem of local optimal solution is also solved. At the same time, the system operation reduces the number of clustering and iterations and improves the efficiency of the algorithm. In addition, this article uses an example to perform algorithm performance analysis. The results show that the proposed algorithm has certain effects and can provide theoretical reference for subsequent related research.