SOMnia constructs
Self-Organizing Maps. It employs Teuvo Kohonen's competitive learning
algorithm to approximate high-dimensional data with a two-dimensional
grid ('map').
For an introduction see
Hertz, J., Krogh, A., Palmer, R. G. (1991), Introduction to the
theory of neural computation, Reading, MA: Addison-Wesley, pp.
217-228, 232-246. Sample applications are
reported in Mazanec, J. A. (1995), 'Competition among European
Tourist Cities: A Comparative Analysis with Multidimensional Scaling
and Self-Organizing Maps', Tourism Economics, Vol. 1, Nr. 3, pp.
283-302 and in Mazanec, J. A. (1995), 'Positioning Analysis with
Self-Organizing Maps, An Exploratory Study on Luxury Hotels', The
Cornell Hotel & Restaurant Administration Quarterly, Vol. 36, Nr.
6, pp. 80-95. The data used in these analyses are available in the
test data files CITY468.DAT (26 cities * 18 years = 468 cases) and
HOTEL152.DAT (4 hotels * 38 respondents = 152 cases).
Note:
You have to read and accept our download
conditions before! File Version OS Location Size SOMnia for DOS 1.22 DOS 50 KB SOMnia for Windows 1.51 Win 3.1 with Win32s or higher 145 KB Testdata 55 KB city468.dat 61 KB hotel152.dat 11 KB Win32s Extension Win 3.1 1.1 MB (Disk 1)
SOMnia for DOS/Windows
26 cities * 18 years
4 hotels * 38 respondents
1.3 MB (Disk 2)