Optimized clustering method

Abstract

The invention discloses an optimized clustering method, and the method specifically comprises the following steps of S1, selecting the pixels in a data set, and building a dense point set Y; S2, selecting the pixel points from the dense point set Y to form a set Q; S3, selecting m pixel points from the data set, and establishing an alternative initial clustering center point set C; S4, dividing the pixel points in the dense point set Y into a class where each initial clustering center in the set Q is located, and obtaining the average maximum similarity of the first clustering; S5, obtaining the minimum cluster average maximum similarity; and S6, taking the clustering center in the set Q corresponding to the minimum clustering average maximum similarity as an initial clustering center of the optimal kmeans clustering, carrying out kmeans clustering, and obtaining a clustering result. In order to reduce the interference of the noise on the data, the density sparse points are eliminatedby using a density distribution function, some noise interference points and abnormal points are eliminated, and the optimal initial clustering center is selected, so that the number k value of the optimal clustering is determined, and the clustering precision is improved.

Description

technical field [0001] The invention relates to the technical field of signal and information processing, in particular to an optimized clustering method. Background technique [0002] With the development of artificial intelligence and the Internet, it has become easier to obtain large-scale data, and the rapid development of various data platforms has gradually laid the foundation for contemporary big data applications. At the same time, in the process of preliminary processing of a large amount of data, it is often required to classify some similar data, and clustering is one of the common techniques for data processing using the distribution characteristics of data. Clustering is a type of unsupervised learning that groups similar objects into the same cluster. The clustering method can be applied to almost all objects, the more similar the objects in the cluster, the better the clustering effect. [0003] Kmeans algorithm is a well-known clustering algorithm, because ...

AI Technical Summary

Problems solved by technology

[0004] Purpose of the invention: In view of the problem that the optimal clustering number k value is not easy to select in the existing clustering methods, the present invention proposes an optimized clustering method

Images