Self-organizing mapping weight particle swarm mean value clustering method, device, equipment and storage medium

Abstract

The embodiment of the invention discloses a self-organizing mapping weight particle swarm mean value clustering method, a device, equipment and a storage medium. According to the method, original sample data can be clustered, firstly, an SOM clustering algorithm is used for conducting coarse clustering on the original sample data, K coarse clustering clusters and coarse clustering centers can be obtained, and therefore the clustering cluster number and the initial clustering center can be determined; and then fine clustering is performed on the original sample data based on the determined coarse clustering cluster and coarse clustering center to obtain a target clustering center, and the initial clustering cluster number and clustering center of fine clustering are determined, so that compared with the condition that the initial value of a traditional clustering algorithm is random, the clustering accuracy and the clustering effect can be improved, the problem that a traditional clustering algorithm is sensitive to an initial value is solved, samples do not need to be marked during clustering, the method can be suitable for clustering of all samples, and the applicability to the samples can be improved.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of data mining, and in particular to a self-organizing map weighted particle swarm mean clustering method, device, equipment and storage medium. Background technique [0002] As an important branch of data mining, clustering algorithms have been widely used in many fields, including machine learning, pattern recognition, image analysis, information retrieval, computer vision, etc. Efficient clustering algorithms can improve work efficiency and work quality . [0003] However, there are some problems in the traditional clustering algorithm. First, the selection of the initial value is random. It is difficult to determine the number of clusters and the initial cluster center, which will directly affect the final clustering result and the accuracy of the clustering. Moreover, different clustering algorithms are suitable for specific data, and no clustering algorithm can be applied to ...

AI Technical Summary

Problems solved by technology

[0003] However, there are some problems in the traditional clustering algorithm. First, the selection of the initial value is random. It is difficult to determine the number of clusters and the initial cluster center, which will directly affect the final clustering result and the accuracy of the clustering.

And different clustering algorithms are suitable for specific data, no clustering algorithm can be applied to all data, the applicability is not good

Images