Fuzzy C-means clustering method based on optimization of Sashimi swarm algorithm

Abstract

The invention discloses a fuzzy C-means clustering method based on optimization of a Sashimi group algorithm, and the method comprises the steps: initializing parameters of the optimization algorithm,and preprocessing a to-be-clustered data set; constructing a target function, namely an evaluation function for a feasible solution; solving the optimal value of the target function to obtain the optimal value and determine the corresponding optimal initial clustering center; and performing fuzzy C-means clustering according to the optimal initial clustering center, and finally obtaining a clustering result. By adopting the method or the system provided by the invention, initial clustering center optimization based on the Saussurea algorithm is carried out on fuzzy C-means clustering, and theproblem that the fuzzy clustering effect is extremely easily influenced by a random initial clustering center is solved.

Description

technical field [0001] The invention relates to the field of clustering methods, in particular to a fuzzy C-means clustering method based on salp swarm algorithm optimization. Background technique [0002] Fuzzy C-means (FCM) is a commonly used clustering method based on an unsupervised learning mechanism, which utilizes a cluster center (distance) calculation function and an FCM objective function. The main steps of FCM are the iterative process, which updates the membership function values ​​and the center locations and their values. In FCM, a high membership value indicates closer to the center of the class, and a low membership value indicates farther away from the center of the class. FCM is extremely sensitive to the initial clustering center. Factors such as the number of clusters, data set characteristics and fuzzy index m all have a great influence on the clustering effect. Different initial clustering centers may lead to different clustering structures. The cluste...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a kind of fuzzy C-means clustering method based on salp group algorithm optimization, carry out the initial clustering center optimization based on salp group algorithm to fuzzy C-means clustering, solve the fuzzy clustering result is very easily affected by Due to the influence of the selection of the initial cluster center, it falls into the problem of local optimum

Images