ASSOM | Adaptive-subspace self-organizing map |

DSS | Decision support system |

EEG | Electroencephalogram |

EM | Expectation maximization |

GNP | Gross national product |

GTM | Generative topographic mapping |

IR | Information retrieval |

KDD | Knowledge discovery in databases |

MDS | Multidimensional scaling |

PCA | Principal component analysis |

RBF | Radial basis function |

SOM | Self-organizing map |

, | input vector (data item), kth input vector |

t | discrete time index |

N | number of input vectors |

input space; n-dimensional Euclidean space | |

ith cluster centroid, ith model vector | |

index of the centroid (or model vector) | |

that is closest to | |

K | number of cluster centroids (and reference vectors) |

projection of | |

d(k,l) | distance between and |

d'(k,l) | distance between and |

cost function of the metric MDS method | |

cost function of the nonmetric MDS method | |

f | a monotonically increasing function used in |

nonmetric MDS | |

cost function of Sammon's mapping | |

q | inherent dimensionality of the data |

neighborhood kernel in the SOM algorithm | |

location of the ith map unit on the map grid | |

probability density function of | |

Voronoi-region corresponding to , viz. the set | |

consisting of those x for which | |

number of data items in | |

centroid of | |

computational complexity (``of the order of'') | |

a modified cost function of the metric MDS method | |

F | a decreasing weighting function |

Mon Mar 31 23:43:35 EET DST 1997