Optimal parameter selection of an svm model: Application to hip fracture risk prediction

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This article presents a study of three (cross) validation metrics used for the selection of the optimal parameters of a support vector machine (SVM) classifier. The study focuses on problems for which the data is non-separable and unbalanced as is often the case for experimental and clinical data. The three metrics selected in this work are the area under the ROC curve, accuracy, and balanced accuracy. As a test example, the study investigates the optimal parameters for an SVM classification model for hip fracture. The hip fracture data is obtained from a finite element model that is fully parameterized. Because the data is computational, fully separable sets of data (fracture and safe) can be obtained. By projection onto a lower dimensional sub-space, the data becomes non-separable and is used to construct the SVM. The knowledge of the separable case provides a comparison metric (the weighted likelihood) that would be unknown if only clinical data is used. The performance of the various metrics are compared for several levels of separability, unbalance and size of training samples. A probabilistic SVM is used to compute the probability of fracture.

Original languageEnglish (US)
Title of host publicationCollection of Technical Papers - AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference
StatePublished - 2013
Event54th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference - Boston, MA, United States
Duration: Apr 8 2013Apr 11 2013

Other

Other54th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference
CountryUnited States
CityBoston, MA
Period4/8/134/11/13

Fingerprint

Support vector machines
Classifiers

ASJC Scopus subject areas

  • Mechanics of Materials
  • Mechanical Engineering
  • Materials Science(all)
  • Aerospace Engineering
  • Architecture

Cite this

Jiang, P., Missoum, S., Hu, C., & Chen, Z. (2013). Optimal parameter selection of an svm model: Application to hip fracture risk prediction. In Collection of Technical Papers - AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference [AIAA 2013-1935]

Optimal parameter selection of an svm model : Application to hip fracture risk prediction. / Jiang, Peng; Missoum, Samy; Hu, Chengcheng; Chen, Zhao.

Collection of Technical Papers - AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference. 2013. AIAA 2013-1935.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Jiang, P, Missoum, S, Hu, C & Chen, Z 2013, Optimal parameter selection of an svm model: Application to hip fracture risk prediction. in Collection of Technical Papers - AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference., AIAA 2013-1935, 54th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference, Boston, MA, United States, 4/8/13.
Jiang P, Missoum S, Hu C, Chen Z. Optimal parameter selection of an svm model: Application to hip fracture risk prediction. In Collection of Technical Papers - AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference. 2013. AIAA 2013-1935
Jiang, Peng ; Missoum, Samy ; Hu, Chengcheng ; Chen, Zhao. / Optimal parameter selection of an svm model : Application to hip fracture risk prediction. Collection of Technical Papers - AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference. 2013.
@inproceedings{67c9c3c6ad6f4ab2bcd6fe0dded1d9f9,
title = "Optimal parameter selection of an svm model: Application to hip fracture risk prediction",
abstract = "This article presents a study of three (cross) validation metrics used for the selection of the optimal parameters of a support vector machine (SVM) classifier. The study focuses on problems for which the data is non-separable and unbalanced as is often the case for experimental and clinical data. The three metrics selected in this work are the area under the ROC curve, accuracy, and balanced accuracy. As a test example, the study investigates the optimal parameters for an SVM classification model for hip fracture. The hip fracture data is obtained from a finite element model that is fully parameterized. Because the data is computational, fully separable sets of data (fracture and safe) can be obtained. By projection onto a lower dimensional sub-space, the data becomes non-separable and is used to construct the SVM. The knowledge of the separable case provides a comparison metric (the weighted likelihood) that would be unknown if only clinical data is used. The performance of the various metrics are compared for several levels of separability, unbalance and size of training samples. A probabilistic SVM is used to compute the probability of fracture.",
author = "Peng Jiang and Samy Missoum and Chengcheng Hu and Zhao Chen",
year = "2013",
language = "English (US)",
isbn = "9781624102233",
booktitle = "Collection of Technical Papers - AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference",

}

TY - GEN

T1 - Optimal parameter selection of an svm model

T2 - Application to hip fracture risk prediction

AU - Jiang, Peng

AU - Missoum, Samy

AU - Hu, Chengcheng

AU - Chen, Zhao

PY - 2013

Y1 - 2013

N2 - This article presents a study of three (cross) validation metrics used for the selection of the optimal parameters of a support vector machine (SVM) classifier. The study focuses on problems for which the data is non-separable and unbalanced as is often the case for experimental and clinical data. The three metrics selected in this work are the area under the ROC curve, accuracy, and balanced accuracy. As a test example, the study investigates the optimal parameters for an SVM classification model for hip fracture. The hip fracture data is obtained from a finite element model that is fully parameterized. Because the data is computational, fully separable sets of data (fracture and safe) can be obtained. By projection onto a lower dimensional sub-space, the data becomes non-separable and is used to construct the SVM. The knowledge of the separable case provides a comparison metric (the weighted likelihood) that would be unknown if only clinical data is used. The performance of the various metrics are compared for several levels of separability, unbalance and size of training samples. A probabilistic SVM is used to compute the probability of fracture.

AB - This article presents a study of three (cross) validation metrics used for the selection of the optimal parameters of a support vector machine (SVM) classifier. The study focuses on problems for which the data is non-separable and unbalanced as is often the case for experimental and clinical data. The three metrics selected in this work are the area under the ROC curve, accuracy, and balanced accuracy. As a test example, the study investigates the optimal parameters for an SVM classification model for hip fracture. The hip fracture data is obtained from a finite element model that is fully parameterized. Because the data is computational, fully separable sets of data (fracture and safe) can be obtained. By projection onto a lower dimensional sub-space, the data becomes non-separable and is used to construct the SVM. The knowledge of the separable case provides a comparison metric (the weighted likelihood) that would be unknown if only clinical data is used. The performance of the various metrics are compared for several levels of separability, unbalance and size of training samples. A probabilistic SVM is used to compute the probability of fracture.

UR - http://www.scopus.com/inward/record.url?scp=84881361365&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84881361365&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84881361365

SN - 9781624102233

BT - Collection of Technical Papers - AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference

ER -