Simulation of Segmented Clustering of Cloud Storage Data Based on Neural Network Models and Python

Abstract

In order to improve the operational efficiency of traditional cloud storage data segmentation clustering methods, the author proposes a machine learning based cloud storage data segmentation clustering method. Reasonably extract multiple small datasets from the cloud storage database, which contain all natural clusters in the cloud storage database. Construct a similarity matrix based on the definition of similarity. Using nonlinear kernel principal component algorithm to measure the similarity of data in the similarity matrix, data with the same features are grouped together through similarity measurement, and a mixed Gaussian distribution probability density model is used to calculate the posterior probability of different categories of data, implement segmented clustering of cloud storage data by comparing probability sizes. The experimental results show that the proposed method can shorten the clustering running time, reduce the clustering variation to 29%, and effectively improve the smoothness of the clustering results.