A potential energy entropy-based Laplace centralized peak data clustering method

Abstract

A potential energy entropy-based Laplace centralized peak data clustering method, including: preprocessing the data set to be categorized, so that the data set to be classified is transformed into a weighted fully coupled network, calculating the Laplacian centrality and the minimum distance of all nodes in the network, calculating the potential energy of any node in the network, calculating the potential energy entropy and automatically extracting the parameters from the data set, and classifying the data set and clustering using the DBSCAN framework. The invention is characterized in that the required parameters are extracted from the original data set, the correct number of clusters is automatically found, and the true parameterless clustering is realized. This will achieve higher accuracy, do not set parameters artificially, automatically complete the effect of the clustering process.

Description

technical field [0001] The invention relates to the field of data mining, in particular to a Laplacian centrality peak data clustering method based on potential energy entropy. Background technique [0002] With the development of science and technology and the diversification of people's means of obtaining data, the amount and structure of data possessed by human beings has been greatly improved. How to mine useful information from these data has increasingly become a necessary technology. . Traditional data analysis is to access and simply operate the data stored in the database. The amount of information contained in the data we obtain through this method is only a small part of the amount of information contained in the entire database. Hidden in these The more important information behind the data is the description of the overall characteristics of the data and the prediction of its development trend, which has important reference value in the process of decision-maki...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the problems that existing clustering algorithms need to manually set parameters and cannot automatically complete clustering in the clustering process, and at the same time consider improving the performance of clustering effects, the present invention proposes a method with high accuracy, no parameters, and can A Laplacian centrality peak data clustering method based on potential energy entropy to automatically complete the clustering process

Images