Multi-Objective Approach for Support Vector Machine Parameter Optimization and Variable Selection in Cardiovascular Predictive Modeling

Christina Brester, Ivan Ryzhikov, Tomi-Pekka Tuomainen, Ari Voutilainen, Eugene Semenkin, Mikko Kolehmainen

Abstract

We present a heuristic-based approach for Support Vector Machine (SVM) parameter optimization and variable selection using a real-valued cooperative Multi-Objective Evolutionary Algorithm (MOEA). Due to the possibility to optimize several criteria simultaneously, we aim to maximize the SVM performance as well as minimize the number of input variables. The second criterion is important especially if obtaining new observations for the training data is expensive. In the field of epidemiology, additional model inputs mean more clinical tests and higher costs. Moreover, variable selection should lead to performance improvement of the model used. Therefore, to train an accurate model predicting cardiovascular diseases, we decided to take a SVM model, optimize its meta and kernel function parameters on a true population cohort variable set. The proposed approach was tested on the Kuopio Ischemic Heart Disease database, which is one of the most extensively characterized epidemiological databases. In our experiment, we made predictions on incidents of cardiovascular diseases with the prediction horizon of 7–9 years and found that use of MOEA improved model performance from 66.8% to 70.5% and reduced the number of inputs from 81 to about 58, as compared to the SVM model with default parameter values on the full set of variables.

Download


Paper Citation


in Harvard Style

Brester C., Ryzhikov I., Tuomainen T., Voutilainen A., Semenkin E. and Kolehmainen M. (2018). Multi-Objective Approach for Support Vector Machine Parameter Optimization and Variable Selection in Cardiovascular Predictive Modeling.In Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO, ISBN 978-989-758-321-6, pages 199-205. DOI: 10.5220/0006866001990205


in Bibtex Style

@conference{icinco18,
author={Christina Brester and Ivan Ryzhikov and Tomi-Pekka Tuomainen and Ari Voutilainen and Eugene Semenkin and Mikko Kolehmainen},
title={Multi-Objective Approach for Support Vector Machine Parameter Optimization and Variable Selection in Cardiovascular Predictive Modeling},
booktitle={Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,},
year={2018},
pages={199-205},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006866001990205},
isbn={978-989-758-321-6},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,
TI - Multi-Objective Approach for Support Vector Machine Parameter Optimization and Variable Selection in Cardiovascular Predictive Modeling
SN - 978-989-758-321-6
AU - Brester C.
AU - Ryzhikov I.
AU - Tuomainen T.
AU - Voutilainen A.
AU - Semenkin E.
AU - Kolehmainen M.
PY - 2018
SP - 199
EP - 205
DO - 10.5220/0006866001990205