Electric power communication data clustering method based on improved K-means algorithm

Abstract

The invention discloses an electric power communication data clustering method based on an improved K-means algorithm. The electric power communication data clustering method comprises the steps of S101, performing normalization processing on electric power communication data; S102, manually selecting an initial classification number K of the normalized data, determining an element distance matrixaccording to the K value, and determining K initial clustering centers; S103, selecting an element, and determining a classification group corresponding to the element by calculating the distance between the element and each initial clustering center; S104, updating the clustering center of each classification group, and determining the actual clustering center of each classification group; S105,obtaining the classification of the power communication data until the classification group does not change any more; according to the method, on the basis of a traditional K-means clustering algorithm, the initial classification number K value can be dynamically adjusted and improved according to the clustering effect so as to improve the clustering effect; initial elements can be selected morereasonably according to the element distance matrix, so that the classification rationality is improved, and the practicability is extremely high.

Description

technical field [0001] The invention belongs to the technical field of electric power communication, and in particular relates to an electric power communication data clustering method based on an improved K-means algorithm. Background technique [0002] There is a huge amount of redundant data in the power communication network. Redundant data processing is an important part of power communication data governance, and data clustering is the pre-process of redundant data processing. Its purpose is to classify the huge power communication data. Therefore, according to the actual situation of various types of data, the types of redundant data are analyzed, and redundant data processing methods are adopted according to local conditions. [0003] The K-means algorithm is the main method for data clustering of the current power communication network. The implementation process of the traditional K-means algorithm is as follows: figure 1 As shown, its main processes include: [...

AI Technical Summary

Problems solved by technology

[0007]K value and initial elements are the key factors to realize element clustering in K-means algorithm. In traditional K-means algorithm, K value and initial elements are given manually It is difficult to guarantee the clustering effect due to the lack of scientific support

Images