The fuzzy set was proposed by Zadeh (1965) in 1965, it is an interesting extension of classical set. In classical sets the characteristic function can take only one of two values from the {0, 1} , in fuzzy sets the membership function can take any value from the interval [0, 1] . The membership function plays an important role in fuzzy set theory and its applications. But in some actual problems, it is often very difficult to determine the value of membership function (or membership degree for short). To solve this problem, the concept of type-2 fuzzy set was introduced by Zadeh (1975) in 1975. A type-2 fuzzy set is characterized by a special membership function, the value of which for each element of this set is a fuzzy set in [0, 1] . Compared with ordinary fuzzy set, type-2 fuzzy set could describe the objective phenomenon more accurately because of secondary membership function. It has been applied more efficiently to pattern recognition and other machine learning fields successfully (Mitchell, 2005).

Support vector machine built on statistical learning theory is a kind of effective machine learning method, has advantaged superiority in dealing with small samples classification problems, and has become the standard tool in machine learning field now (Cristianini and Shawe-Taylor, 2000). The traditional support vector machine is based on the real valued random samples and established on the probability space, and it is difficult to handle the small samples classification problems with non-real random samples on non-probability space. Naturally, it is a very interesting and valuable research direc-tion to construct the support vector machine based on non-real valued random samples and established on nonprobability space or probability space. On this basis, Lin and Wang (2002) proposed the support vector machine based on a non-real valued random samples-fuzzy samples, constructed the fuzzy support vector machine, which determines membership degree by different weight for each sample. Then Ha et al. (2009, 2011) improved the fuzzy support vector machine in determining membership degree, and they constituted a new fuzzy support vector machine and an intuitionistic fuzzy support vector machine. Considering that above support vector machines only make the samples fuzzified and the training samples are not fuzzy input directly, Ji et al. (2010) constructed a kind of support vector machine for classification based on fuzzy training data, and discussed its application in the diagnosis of coronary. Nevertheless the support vector machine is limited to the ordinary fuzzy number samples-the triangular fuzzy samples, and it is difficult to deal with the classification problems of type- 2 fuzzy number training samples which is a generalization of the ordinary fuzzy number samples. In order to handle the problem, this paper firstly discusses the support vector machine based on type-2 fuzzy training samples and established on generalized credibility space.

Definition 2.1(Zadeh, 1965): A fuzzy set A in X is characterized by a mapping μ_A : X → [0, 1], χ ？ μ_A(χ), μ_A is called the membership function of A and μ_A(χ) is called the membership degree of χ in A. The symbol

represents A. A collection of all fuzzy sets of X can be denoted by

Definition 2.2: Suppose a, b, c ∈ R. a is a fuzzy set in R, and its membership function is as follows

Then a is called triangular fuzzy number, and is represented by a = (a, b, c) (refer to Ha et al., 2010)

Definition 2.3(Zadeh, 1975): A type-2 fuzzy set in X , denoted

is characterized by a secondary membership function

where χ ∈ X, u ∈ J _x ⊆ [0,1], and

can also be expressed as

Definition 2.4 (Qin et al., 2011): A type-2 fuzzy set

is called triangular if its secondary membership function

is

for x ∈ [r₁, r₂], and

for x ∈ [r₂ , r₃], where θ₁, θ₂ ∈ [0, 1] are two parameters characterizing the degree of uncertainty that

takes the value x. For simplicity, we denote the type-2 triangular fuzzy number

with the above condition by (r₁, r₂, r₃ ; θ₁, θ_r).

Definition 2.5(Qin et al., 2011): Let X be a nonempty set, P( X ) be the class of all subsets of X , Pos be a possibility measure in P(X). Suppose a is a fuzzy number, then the possibility measure of fuzzy event a ≤ b is defined by Pos({a ≤ b} ) = sup {μ_a(χ), χ ∈ R, χ ≤ b} . Similarly, Pos({ a < b} ) = sup{μ_a(χ), χ ∈ R, χ < b}, Pos({a = b}) = μ_a(b).

Definition 2.6(Qin et al., 2011): Let A be a fuzzy set in X, and

be a type-2 fuzzy set in X. A is the reduction of

via the mean reduction method, if it satisfies

We denote

Example 2.1(Qin et al., 2011): Let

be a type-2 triangular fuzzy number. According to the definition 2.6, we have the reduction a of

and the membership function of a is as follows.

Definition 2.7(Qin et al., 2011): Suppose a is a fuzzy number. The generalized credibility measure

of the event {a ≤ b} is defined by

Theorem 2.1(Qin et al., 2011): Let a_i be the mean reduction of type-2 fuzzy triangular number

i = 1, 2, … , n. Suppose a₁, a ₂, … ,a_n are mutually independent, and k_i ≥ 0, i = 1, 2, … , n. Given the generalized credibility level

is equivalent to

Proof: It can be easily got by type-2 fuzzy set theory in the reference.

Consider the fuzzy type-2 training samples set

where

and

i = 1, 2, …, m, j = 1,2 … , n, when y_i = 1, then

is called a positive class; when y_i = -1, then

is called a negative class. The classification based on the type-2 fuzzy training set

is to find a decision function

such that the positive class and the negative class can be separated with the low classification error and good generalization performance.

Definition 3.1(Ji et al., 2010): For the type-2 fuzzy training samples set

if for a given level

there exist w∈ Rⁿ, b∈R, such that

Then the type-2 fuzzy training samples set

is strong type-2 fuzzy linear separable.

The support vector machine for strong linear separable type-2 fuzzy sample set is to solve the fuzzy chance constrained programming:

To solve the above programming,

is equivalent to

i = 1, 2, …, m, j = 1, 2, … , n according to Theorem 2.1. Then the fuzzy chance constrained programming is converted into the following classic convex quadratic programming:

We can obtain its dual problem:

w^* , b^* are the solution of above dual programming. For a given confidence level

if

then

belongs to the positive class; if

then

belongs to the negative class.

We shall apply the support vector machine for twoclass classification with type-2 fuzzy training samples to the diagnosis of Coronary. The data is sourced from the reference (Ji et al., 2010), the half of which are healthy (y_i = 1), the others are Coronary patients (y_i = -1). Let θ_l,1ⁱ = θ _r,1ⁱ = 0.1, θ_l,2ⁱ = θ_r,2ⁱ = 0.2, we can obtain type-2 triangular fuzzy numbers

and

The programming was solved by SVM toolbox of matlab, when parameter C =1, λ = 0.8. The accurate rate of diagnosis is 80% . The result of example illustrates that the proposed support vector machine is effective.

This paper constructs firstly the support vector machine in which the training samples is type-2 fuzzy number input, and the proposed support vector machine is an interesting extension of the traditional support vector machine and the fuzzy support vector machine. We shall focus on the proposed support vector machine’s applications in some practical problems.