Title: Improving the predictive ability of multivariate calibration models using support vector data description

Authors: Walid Gani

Addresses: Tunisian Competition Council, Rue du Lac Biwa, Les Berges du Lac, 1053 Tunis, Tunisia

Abstract: Outlier detection is a crucial step in building multivariate calibration models and enhancing their predictive ability. However, traditional outlier detection methods often suffer from important drawbacks mainly their reliance on assumptions about the data model distribution and their unsuitability for real-life applications. This paper investigates the use of support vector data description (SVDD) for the detection of outliers and proposes a multivariate calibration strategy that combines partial least squares (PLS) and SVDD. For the assessment of the proposed calibration strategy, an experimental study aiming to predict four properties of diesel fuel is conducted. The results show that the predictive ability of PLS-SVDD is better than the predictive ability of a classical strategy that combines PLS and the T2 method.

Keywords: multivariate calibration? outlier? support vector data description? SVDD? partial least squares? PLS? T2 method.

DOI: 10.1504/IJDATS.2021.118021

International Journal of Data Analysis Techniques and Strategies, 2021 Vol.13 No.3, pp.227 - 243

Received: 25 Mar 2019
Accepted: 23 Mar 2020

Published online: 08 Oct 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article