Journal of Statistics Applications & Probability
Abstract
The support vector machine (SVM) is widely used for machine learning and artificial intelligence. Traditional support vector machine has been extended to multicategory case for multicategory classification problem. However, it does not provide an established Bayesian Framework for Multicategory Support Vector Machine. Corresponding to this, we propose Bayesian methods for multi-class support vector machine. Extensive numerical studies were conducted to evaluate the performance of the proposed method. The numerical study suggests that the proposed Bayesian framework provides good results for practical situations. In addition, an illustrative example using MIT Genome data is presented.
Digital Object Identifier (DOI)
http://dx.doi.org/10.18576/jsap/090101
Recommended Citation
Liu, Yeqian
(2020)
"Extended Bayesian Framework for Multicategory Support Vector Machine,"
Journal of Statistics Applications & Probability: Vol. 9:
Iss.
1, Article 1.
DOI: http://dx.doi.org/10.18576/jsap/090101
Available at:
https://digitalcommons.aaru.edu.jo/jsap/vol9/iss1/1