Improved research based on density peak clustering algorithm

Abstract

The invention relates to an improved research of a clustering algorithm based on a density peak value, belongs to one of clustering algorithms, belongs to unsupervised classification, and aims to divide data into different clusters. The density peak clustering algorithm determines a clustering center according to the decision diagram and detects non-spherical clusters without specifying the number of clusters. The invention aims to solve the problems existing in a traditional DPC clustering algorithm, the traditional DPC algorithm processes data, calculates local density and minimum distance, constructs a decision diagram through the local density and the minimum distance, and manually selects a point with relatively large local density and minimum distance as a clustering center point, so that the clustering accuracy is not high, and the clustering accuracy is high. Therefore, aiming at the problems that a density peak value clustering algorithm cannot adaptively select a threshold value, a domino effect is easily generated by distributing residual points and the like, the DTW algorithm is introduced, the adaptive threshold value is designed, and the DPC clustering algorithm is improved, so that the defects existing in the DPC clustering algorithm are improved, and the clustering accuracy is improved.

Description

technical field [0001] The invention relates to the application field of computer technology, in particular to an improved research on a density peak-based clustering algorithm. Background technique [0002] Clustering analysis is an unsupervised method to analyze the relationship between data, and it is a preprocessing step of data mining. Clustering analysis is the process of dividing data into groups. Physical objects are divided into several classes. The similarity between objects within a class is high, and the similarity between objects between classes is low. Cluster analysis is widely used in the fields of data mining, gene identification, image processing and document retrieval. [0003] Traditional clustering algorithms can be roughly divided into partition clustering methods, hierarchical clustering methods, density clustering methods, grid clustering methods, and model clustering methods. The two classical algorithms in the partition-based clustering algorithm...

AI Technical Summary

Problems solved by technology

[0006] In view of this, the present invention mainly solves the problem that the density peak clustering algorithm cannot adaptively select the threshold, and avoids the domino effect caused by the allocation of remaining points. Once a certain sample is allocated incorrectly, it will lead to the problem of subsequent sample allocation errors.

Images