Asian Science Citation Index is committed to provide an authoritative, trusted and significant information by the coverage of the most important and influential journals to meet the needs of the global scientific community.  
ASCI Database
308-Lasani Town,
Sargodha Road,
Faisalabad, Pakistan
Fax: +92-41-8815544
Contact Via Web
Suggest a Journal
Information Technology Journal
Year: 2013  |  Volume: 12  |  Issue: 12  |  Page No.: 2412 - 2418

Flood Disaster Evaluation Model Based on Kernel Dual Optimization Support Vector Machine

Weiping Deng, Jianzhong Zhou, Qiang Zou, Jian Xiao, Yongchuan Zhang and Weihua Hua    

Abstract: Support vector machine is adopted in this paper to construct flood disaster evaluation model, which can be indicated as a comprehensively nonlinear classification issue. In this article, kernel function of SVM is optimized by kernel transformation and kernel parameters optimization. In order to discriminate the flood disaster evaluation indexes and really reflect their classification contributions, kernel function is weighted to promote classification performance and reduce the error influence by weak features. Further more, after analyzing the over learning issue of traditional grid search, an improved grid search is proposed to optimize the kernel parameters. The new search method illustrates it is more reasonable to make second search around some suboptimal solutions. By this search method, the final solution obtains high accuracy in testing samples and reduces the value of penalty factor. The experiment results show the promotion of classification precision by the dual optimization model and identify it could be a good choice for other classification issues.

Cited References   |    Fulltext    |   Related Articles   |   Back
  Related Articles

Copyright   |   Desclaimer   |    Privacy Policy   |   Browsers   |   Accessibility