Year: 2005
Type: article
Abstract: Recent development of Self-Organizing Feature Maps (hereafter SOFMs) algorithm becomes more and more interesting in many fields such as: pattern recognition, clustering, and function approximation, data-and web-mining.This paper will cover issues about the SOFM its self as well as some of its variances like Neural Gas, Growing Neural Gas, Growing Cell Structures, etc and they are all the kind of unsupervised competitive learning algorithms.This paper will be organized into 5 parts.After the introduction to the SOFM algorithm and some discussion on the motivations that led to the origin is given in part 1, in part 2, the goodness as well as the weakness of this model is discussed.In part 3, the details about the variances most noticeable are discussed in part 2. Exploring the variances of the original SOFM algorithm mentioned in that part will do it.Part 4 gives some discussion on the general of the applications of this model in author's opinions.A conclusion in part 5 is given to conclude this paper.
Authors: Duc Minh Tran, Hai Khoi Le
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2565468409