A fuzzy clustering algorithm based on complex synaptic neural network

This paper presents a new fuzzy clustering algorithm to solve the problem that the fuzzy c-means (FCM) clustering algorithm has a poor accuracy of clustering. The adopted methodology used the minimum support tree principle to get the initial clustering center and augmented lagrange multiplier method to solve the problem of noise sensitive of the FCM algorithm. Besides, the Hopfield neural network is used to calculate the cluster center and complex synaptic neural network is used to obtain the membership grades. In the experiment, the proposed algorithm is simulated and compared with the commonly used clustering algorithm. The clustering accuracy is higher than that of other algorithms. The analysis proves that the algorithm has universal guiding significance in theory and engineering practice.