Growth of Hydrocarbon Particles Inside Multidimensional Clustering by Gaseous Neural Network Method

Abstract

Gas neural network or neural gas network is one of the types of competitive neural networks with unsupervised learning pattern, whose main application is in solving clustering problems and learning topology. This type of neural network, in terms of classification, falls into the category of Vector Quantification algorithms (VQ) and is very closely related to the k-Means clustering algorithm, SOM neural network (self-organizing maps) and LVQ neural network. Purpose: The purpose of this article is that we should be able to convert very large and huge data into very small and useful data. has it. In addition to clustering and placing the center of the clusters in the right place, the neural gas network dynamically creates neighborhood connections between neurons (cluster centers), which ultimately enables this algorithm to learn topology.