Multiobjective Optimization of Classifiers by Means of 3-D Convex Hull Based Evolutionary Algorithm

Created by W.Langdon from gp-bibliography.bib Revision:1.4221

  title =        "Multiobjective Optimization of Classifiers by Means of
                 {3-D} Convex Hull Based Evolutionary Algorithm",
  note =         "Comment: 32 pages, 26 figures",
  author =       "Jiaqi Zhao and Vitor Basto Fernandes and 
                 Licheng Jiao and Iryna Yevseyeva and Asep Maulana and Rui Li and 
                 Thomas Baeck and Michael T. M. Emmerich",
  year =         "2014",
  month =        dec # "~17",
  bibsource =    "OAI-PMH server at",
  oai =          "",
  keywords =     "genetic algorithms, genetic programming",
  URL =          "",
  abstract =     "Finding a good classifier is a multiobjective
                 optimisation problem with different error rates and the
                 costs to be minimised. The receiver operating
                 characteristic is widely used in the machine learning
                 community to analyse the performance of parametric
                 classifiers or sets of Pareto optimal classifiers. In
                 order to directly compare two sets of classifiers the
                 area (or volume) under the convex hull can be used as a
                 scalar indicator for the performance of a set of
                 classifiers in receiver operating characteristic space.
                 Recently, the convex hull based multiobjective genetic
                 programming algorithm was proposed and successfully
                 applied to maximise the convex hull area for binary
                 classification problems. The contribution of this paper
                 is to extend this algorithm for dealing with higher
                 dimensional problem formulations. In particular, we
                 discuss problems where parsimony (or classifier
                 complexity) is stated as a third objective and
                 multi-class classification with three different true
                 classification rates to be maximised. The design of the
                 algorithm proposed in this paper is inspired by
                 indicator-based evolutionary algorithms, where first a
                 performance indicator for a solution set is established
                 and then a selection operator is designed that complies
                 with the performance indicator. In this case, the
                 performance indicator will be the volume under the
                 convex hull. The algorithm is tested and analysed in a
                 proof of concept study on different benchmarks that are
                 designed for measuring its capability to capture
                 relevant parts of a convex hull. Further benchmark and
                 application studies on email classification and feature
                 selection round up the analysis and assess robustness
                 and usefulness of the new algorithm in real world
  notes =        "see \cite{Zhao:2016:IS}",

Genetic Programming entries for Jiaqi Zhao Vitor Basto-Fernandes Licheng Jiao Iryna Yevseyeva Asep Maulana Rui Li Thomas Back Michael Emmerich