Classification of Multivariate Objects Using Interval Quantile Classes |
| |
Authors: | Andrzej M?odak |
| |
Institution: | 1. Statistical Office in Pozna??, Branch in Kalisz, pl. J. Kili??skiego 13, 62-800, Kalisz, Poland
|
| |
Abstract: | The paper contains a proposal of interval data clustering related to given social and economic objects characterized by many
interval variables. This multivariate approach is based on an original conception of interval quantiles constructed using
a special definition derived from the notion of the Hausdorff distance. In order to improve the quality of classification,
the obtained interval quantile classes can be next aggregated into larger merged classes. The efficiency of our method can
be assessed using especially defined indices of entropy and volume coefficients. The second notion replaces the classical
concept of area, which is not applicable in this case. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|