Corporate financial crisis prediction using SVM models with direct search for features selection and parameters optimization

Ligang Zhou; Kin Keung Lai

doi:10.1109/CSO.2012.172

Corporate financial crisis prediction using SVM models with direct search for features selection and parameters optimization

Ligang Zhou
, Kin Keung Lai

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

1 Citation (Scopus)

Abstract

since the accuracy of corporate financial crisis prediction is very important for financial institutions, investors and governments, many methods have been employed for developing effective prediction models. Support vector machines (SVM) are powerful methods for classification and have been used for this task. However, the performance of SVM is sensitive to parameters optimization and features selection. In this study, a new approach based on direct search and features ranking technology is proposed to combine features selection and parameters optimization for SVM models for financial crisis prediction. The sensitivity of features ranking technology, strategies of sampling training samples, and types of SVM models are analyzed on a data set with 2010 samples. The experimental results show that the proposed models are good alternatives for financial crisis prediction. © 2012 IEEE.

Original language	English
Title of host publication	Proceedings of the 2012 5th International Joint Conference on Computational Sciences and Optimization, CSO 2012
Pages	760-764
DOIs	https://doi.org/10.1109/CSO.2012.172
Publication status	Published - 2012
Event	2012 5th International Joint Conference on Computational Sciences and Optimization, CSO 2012 - Harbin, Heilongjiang, China Duration: 23 Jun 2012 → 26 Jun 2012

Conference

Conference	2012 5th International Joint Conference on Computational Sciences and Optimization, CSO 2012
Place	China
City	Harbin, Heilongjiang
Period	23/06/12 → 26/06/12

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 10 Reduced Inequalities

Research Keywords

Direct search
Features ranking
Support vector machines

Access Output

10.1109/CSO.2012.172

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{3e8858ac5095445b912b92253a76b848,

title = "Corporate financial crisis prediction using SVM models with direct search for features selection and parameters optimization",

abstract = "since the accuracy of corporate financial crisis prediction is very important for financial institutions, investors and governments, many methods have been employed for developing effective prediction models. Support vector machines (SVM) are powerful methods for classification and have been used for this task. However, the performance of SVM is sensitive to parameters optimization and features selection. In this study, a new approach based on direct search and features ranking technology is proposed to combine features selection and parameters optimization for SVM models for financial crisis prediction. The sensitivity of features ranking technology, strategies of sampling training samples, and types of SVM models are analyzed on a data set with 2010 samples. The experimental results show that the proposed models are good alternatives for financial crisis prediction. {\textcopyright} 2012 IEEE.",

keywords = "Direct search, Features ranking, Support vector machines",

author = "Ligang Zhou and Lai, \{Kin Keung\}",

year = "2012",

doi = "10.1109/CSO.2012.172",

language = "English",

isbn = "9780769546902",

pages = "760--764",

booktitle = "Proceedings of the 2012 5th International Joint Conference on Computational Sciences and Optimization, CSO 2012",

note = "2012 5th International Joint Conference on Computational Sciences and Optimization, CSO 2012 ; Conference date: 23-06-2012 Through 26-06-2012",

}

Zhou, L & Lai, KK 2012, Corporate financial crisis prediction using SVM models with direct search for features selection and parameters optimization. in Proceedings of the 2012 5th International Joint Conference on Computational Sciences and Optimization, CSO 2012., 6274835, pp. 760-764, 2012 5th International Joint Conference on Computational Sciences and Optimization, CSO 2012, Harbin, Heilongjiang, China, 23/06/12. https://doi.org/10.1109/CSO.2012.172

Corporate financial crisis prediction using SVM models with direct search for features selection and parameters optimization. / Zhou, Ligang; Lai, Kin Keung.
Proceedings of the 2012 5th International Joint Conference on Computational Sciences and Optimization, CSO 2012. 2012. p. 760-764 6274835.

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Corporate financial crisis prediction using SVM models with direct search for features selection and parameters optimization

AU - Zhou, Ligang

AU - Lai, Kin Keung

PY - 2012

Y1 - 2012

N2 - since the accuracy of corporate financial crisis prediction is very important for financial institutions, investors and governments, many methods have been employed for developing effective prediction models. Support vector machines (SVM) are powerful methods for classification and have been used for this task. However, the performance of SVM is sensitive to parameters optimization and features selection. In this study, a new approach based on direct search and features ranking technology is proposed to combine features selection and parameters optimization for SVM models for financial crisis prediction. The sensitivity of features ranking technology, strategies of sampling training samples, and types of SVM models are analyzed on a data set with 2010 samples. The experimental results show that the proposed models are good alternatives for financial crisis prediction. © 2012 IEEE.

AB - since the accuracy of corporate financial crisis prediction is very important for financial institutions, investors and governments, many methods have been employed for developing effective prediction models. Support vector machines (SVM) are powerful methods for classification and have been used for this task. However, the performance of SVM is sensitive to parameters optimization and features selection. In this study, a new approach based on direct search and features ranking technology is proposed to combine features selection and parameters optimization for SVM models for financial crisis prediction. The sensitivity of features ranking technology, strategies of sampling training samples, and types of SVM models are analyzed on a data set with 2010 samples. The experimental results show that the proposed models are good alternatives for financial crisis prediction. © 2012 IEEE.

KW - Direct search

KW - Features ranking

KW - Support vector machines

UR - https://www.scopus.com/pages/publications/84867912479

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-84867912479&origin=recordpage

U2 - 10.1109/CSO.2012.172

DO - 10.1109/CSO.2012.172

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9780769546902

SP - 760

EP - 764

BT - Proceedings of the 2012 5th International Joint Conference on Computational Sciences and Optimization, CSO 2012

T2 - 2012 5th International Joint Conference on Computational Sciences and Optimization, CSO 2012

Y2 - 23 June 2012 through 26 June 2012

ER -

Corporate financial crisis prediction using SVM models with direct search for features selection and parameters optimization

Abstract

Conference

UN SDGs

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this