Journalartikel

Jahr der Veröffentlichung: 2020

Zeitschrift: Expert Systems: The Journal of Knowledge Engineering

Bandnummer: 37

Heftnummer: 3

ISSN: 0266-4720

eISSN: 1468-0394

Open Access Status: Hybrid

DOI Link: https://doi.org/10.1111/exsy.12506

Verlag: Wiley

Abstract: 
The key property of monotone classifiers is that increasing (decreasing) input values lead to increasing (decreasing) the output value. Preserving monotonicity for a classifier typically requires many constraints to be respected by modeling approaches such as artificial intelligence techniques. The type of constraints strongly depends on the modeling assumptions. Of course, for sophisticated models such conditions might be very complex. In this study we present a new family of kernels that we call it Choquet kernels. Henceforth it allows for employing popular kernel-based methods such as support vector machines. Instead of a naive approach with exponential computational complexity we propose an equivalent formulation with quadratic time in the number of attributes. Furthermore, since coefficients derived from kernel solutions are not necessarily monotone in the dual form, different approaches are proposed to monotonize coefficients. Finally experiments illustrate beneficial properties of the Choquet kernels.

Harvard-Zitierstil: Tehrani, A., Strickert, M. and Ahrens, D. (2020) Class of Monotone Kernelized Classifiers on the basis of the Choquet Integral, Expert Systems, 37(3), Article e12506. https://doi.org/10.1111/exsy.12506

APA-Zitierstil: Tehrani, A., Strickert, M., & Ahrens, D. (2020). Class of Monotone Kernelized Classifiers on the basis of the Choquet Integral. Expert Systems. 37(3), Article e12506. https://doi.org/10.1111/exsy.12506