Support vector machine to monitor greenhouse plant with Gaussian loss function

Abstract

In this paper, it applys Gaussian loss function instead of ε-insensitive loss function in a standard SVRM to devise a new model and a new type of support vector classification machine whose optimization problem is easier to solve and has conducted effective test on open data set in order to apply the new algorithm to environment monitoring in greenhouse plants and the monitoring result is better than any other method available. © 2011 IFIP International Federation for Information Processing.