Input Selection and Regression using the Som

Abstract

This Paper Presents a Global Methodology to Build a Nonlinear Regression When the Number of Available Samples is Small Compared to the Number of Inputs. the Task is Divided in Two Parts: Selection of the Best Inputs and Construction of the Approximator. a First Som is Used to Compute Clean Correlations between the Inputs and the Output. a Second Som is Built to Link the Output to the Selected Inputs. the Good Performances of This Methodology Are Illustrated on a Spectrometric Dataset.

Department(s)

Engineering Management and Systems Engineering

Keywords and Phrases

Regression; Self-organizing maps; Spectrometry; Variable selection

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2024 WSOM, All rights reserved.

Publication Date

01 Dec 2005

This document is currently not available here.

Share

 
COinS