Cost-Sensitive Support Vector Machine Based on Weighted Attribute

In practice it is existed a matter in classified problem. The problem can be described as that the different sort has different wrong classified cost. In this paper we propose a cost-sensitive SVM approach based on weighted attribute. The approach first calculates the weightiness of feature attributes corresponded to the classification attribute, then calculates the corresponding weightiness of attribute for all sample. In the end the samples are used for cost-sensitive SVM training and testing. The experimental results show that the approach can improve the classification precision of the cost sensitive samples, and also the use of feature attribute increases the integer classified capability of the classifier. The approach has important realistic significance of unbalanced wrong-classification cost in classified problem.