The Investigation and Application of SVC and SVR in Handling Missing Values

The great achievements have been approached in the development of support vector machine (SVM). It has been successfully used for solving classification and regression problems. This paper aims at proposing two algorithms based on SVC and SVR which are two applications of SVM in the fields of classification and regression, to handle both nominal and numerical missing values. Two experiments are conducted. The results indicate that our algorithms provide a high accuracy when compared with some other commonly used algorithms.