论文信息 - Symbolic Linear Regression with Taxonomies

Symbolic Linear Regression with Taxonomies

This work deals with the extension of classical linear regression to symbolic data and constitutes a continuation of previous papers from Billard and Diday on linear regression with interval and histogram-valued data. In this paper, we present a method for regression with taxonomic variables. Taxonomic variables are variables organized in a tree with several levels of generality. For example, the towns are aggregated up to their regions, the regions are aggregated up to their country.

[1] P. Groenen,et al. Data analysis, classification, and related methods , 2000 .

[2] Edwin Diday,et al. Régression linéaire symbolique avec variables taxonomiques , 2004, EGC.

[3] Hans-Hermann Bock,et al. Analysis of Symbolic Data: Exploratory Methods for Extracting Statistical Information from Complex Data , 2000 .

[4] L. Billard,et al. Symbolic Regression Analysis , 2002 .

[5] L. Billard,et al. Regression Analysis for Interval-Valued Data , 2000 .

[6] L. Billard,et al. From the Statistics of Data to the Statistics of Knowledge , 2003 .

[7] Hans-Hermann Bock,et al. Classification, Clustering, and Data Analysis , 2002 .