ePrints

Multivariate decision trees for the interrogation of bioprocess data

Lookup NU author(s): Dr Kathryn Kipling, Professor Gary Montague, Elaine Martin

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.

Abstract

This paper introduces a method for analysing correlated process data to extract useful information with multivariate decision trees. The techniques have been applied to a bioprocess data set and have been shown to provide insight into the causes of process bariation. The differences between the univariate and multivariate methods are highlighted and the tree interpretations are discussed. It is observed that there is very little difference in the classification abilities of the tree types from a prediction of the outcome perspective but the information from the multivariate trees is more informative with regard to cause of deviation rather than simply identifying the observed effects. © 2005 Elsevier B.V. All rights reserved.

Publication metadata

Author(s): Kipling K, Montague G, Martin E, Morris J

Publication type: Article

Publication status: Published

Journal: Computer Aided Chemical Engineering

Year: 2005

Volume: 20

Issue: C

Pages: 1129-1134

ISSN (print): 1570-7946

ISSN (electronic):

Publisher: Elsevier BV

URL: http://dx.doi.org/10.1016/S1570-7946(05)80030-6

DOI: 10.1016/S1570-7946(05)80030-6

Altmetrics

Share