A Linear Programming Model of Multi-class Support Vector Machine
Journal Title: Scholars Journal of Engineering and Technology - Year 2017, Vol 5, Issue 2
Abstract
The structure of K-SVCR algorithm is ‘one-against-one-against-rest’．Its advantage is in the process of each decomposition , make all training points of information have been fully taken advantage of. To a certain extent, can prevent the classification error by incomplete information . But this algorithm constructed a quadratic programming which restricted this algorithm’s speed and the range of applications. So, this paper constructed a linear programming model based on K-SVCR , and then adopt the effective algorithm of Predictor-corrector Method of Mehrotra to solve the linear programming. Preliminary numerical experiments on benchmark datasets show that the algorithm has good performance on both accuracy and training speed than K-class Support Vector Classification-Regression. Keywords: Support vector machine；Multi-class classification；Primal dual interior point method.
Authors and Affiliations
Sa Qi, Yuan Yuping, Tong Ying,, Kan Tao, Zhang Jing
Keywords
Related Articles
- EP ID EP385470
- DOI -
- Views 51
- Downloads 0
Empirical evaluation of dynamic parameters of simple and nano drilling fluid
The energy of a nation is largely conditioned by the extent of prospective, surveyed and extracted oil and gas reserves and other natural resources. Hence this is the main source of somewhere economy. Experiments are hel...
Research on the network transmission of cooperative work in the logging platform of large data environment
Currently advanced methods of logging interpretation method such as acoustic-electric imaging, MRI, array acoustic have been widely applied. Data needed handling are large two-dimensional and even three-dimensional data....
Sharing of Personal Information Using Online Social Network Securely with Stranger Detection Algorithm
Photo sharing is an alluring component which advances Online Social Networks (OSNs). Sadly, it might release clients' protection in the event that they are permitted to post, remark, and label a photo freely .we endeavor...
A New Non-monotone Self-Adaptive Trust Region Method with Fixed Step-size for Unconstrained Optimization
In this paper, we propose and analyze a new non-monotone self-adaptive trust region method with fixed step-size for unconstrained optimization. Unlike the traditional non-monotone trust region method, our algorithm utili...
Failure early warning model for Electrical Submersible Pump based on regression analysis
Take 120 the Electrical Submersible Pump (ESP)well as the object of study, has constructed based on 6 commonly used target return early warning models, then to dives the Electrical Submersible Pump breakdown well to carr...