Article Preview
TopClassification Algorithm Based On Support Vector Machine
Classification algorithm is a kind of data mining technology, which can construct a classification function or model based on characteristics of data collection. Currently, classification algorithm concludes decision tree method, Bayesian method, support vector machine, genetic algorithm and so on. The support vector machine has good generalization ability, and has good nonlinear data processing ability, which has been applied in many fields, such as fault diagnosis, image processing, and text categorization. Therefore, the data mining classification algorithm base on fuzzy support vector machine is applied in customer relationship management (Mo and Zhao, 2016).
The basic idea of fuzzy support vector machine can choose different membership degree according to effect degree of different inputting sample on customer relationship management, a group of samples is given, which is defined by , where, denotes inputting vector of fuzzy support machine; denotes that belongs to one class of two classes, ; denotes the membership degree of class concluding the sample, (Zhang et al., 2017). Fuzzy support vector machine has the same object with the support vector machine, the two classes can be divided based on super plane, and make distance between the supper plane and two classes biggest, the mathematical model of fuzzy support vector machine is expressed as follows (Zhang et al., 2015):Objective function: (1) Boundary condition: , , (2) where, denotes the error measure of support vector machine, denotes the error measure of different weight.