Functional-bandwidth kernel for Support Vector Machine with Functional Data: An alternating optimization algorithm

R. Blanquero, E. Carrizosa, A. Jiménez-cordero, B. Martín-barragán

Research output: Contribution to journalArticlepeer-review

Abstract

Functional Data Analysis (FDA) is devoted to the study of data which are functions. Support Vector Machine (SVM) is a benchmark tool for classification, in particular, of functional data. SVM is frequently used with a kernel (e.g.: Gaussian) which involves a scalar bandwidth parameter. In this paper, we propose to use kernels with functional bandwidths. In this way, accuracy may be improved, and the time intervals critical for classification are identified. Tuning the functional parameters of the new kernel is a challenging task expressed as a continuous optimization problem, solved by means of a heuristic. Our experiments with benchmark data sets show the advantages of using functional parameters and the effectiveness of our approach.
Original languageEnglish
Pages (from-to)195-207
JournalEuropean Journal of Operational Research
Volume275
Issue number1
Early online date24 Nov 2018
DOIs
Publication statusPublished - 16 May 2019

Keywords

  • data mining
  • Functional Data classification
  • parameter tuning
  • SVM
  • functional bandwidth

Fingerprint Dive into the research topics of 'Functional-bandwidth kernel for Support Vector Machine with Functional Data: An alternating optimization algorithm'. Together they form a unique fingerprint.

Cite this