A Multiclass Classification Tool Using Cloud Computing Architecture
Journal
International Symposium on Network Enabled Health Informatics, Biomedicine and Bioinformatics HI-BI-BI 2012
Date Issued
2012-08
Author(s)
Chia-Ping Shen
Chia-Hung Liu
Feng-Sheng Lin
Han Lin
Chi-Ying F. Huang
Cheng-Yan Kao
Feipei Lai
Jeng-Wei Lin
Abstract
Multiclass classification is an important technique to many complex biomedicine problems. Genetic algorithms (GA) are proven to be effective to select features prior to multiclass classification by support vector machines (SVM). However, their use is computation intensive. Based on SOA (Service Oriented Architecture) design principles, this paper proposes a cloud computing framework that exploits the inherent parallelism of GA-SVM classification to speed up the work. The performance evaluations on an mRNA benchmark cancer dataset have shown the effectiveness and efficiency of the framework. With a user-friendly web interface, the framework provides researchers an easy way to investigate the unrevealed secrets in the fast-growing repository of biomedical data.
SDGs
Type
conference paper