Please use this identifier to cite or link to this item:
|Title:||Self-organizing feature maps: the traveling salesman problem and other applications|
Archer, Norman P.
McMaster University, Michael G. DeGroote School of Business
|Series/Report no.:||Research and working paper series (Michael G. DeGroote School of Business)|
|Abstract:||<p>Self-organizing feature maps (SOFM) have received much attention recently. SOFMs are basically a variant of neural network models which use unsupervised learning to acquire and organize their internal structure. Many mathematical features of the model have been discovered. In addition, many applications have been developed. This article is reviews the basic SOFM model as proposed by Kohonen. Next, using the traveling salesmen problem (TSP) as a benchmark, a few variants of the SOFM proposed to solve the TSP are compared to a variety of other algorithms such as the Hopfield/Tank network and simulated annealing. Brief descriptions of all algorithms are included. Lastly, a number of applications of the SOFM are discussed, such as speech and semantic recognition. The objective of this paper is to offer the reader some insight into the SOFM as well as some guidance as to further research.</p>|
|Description:||<p>39 leaves : ; Includes bibliographical references (leaves 36-39). ; "October, 1993".</p>|
|Appears in Collections:||DeGroote School of Business Working Paper Series|
Items in MacSphere are protected by copyright, with all rights reserved, unless otherwise indicated.