2. A linear classier in SVM can be mathematically expressed as f(x) = iwii(x), where each i(x) is a feature function of the original feature set x. Note that the transformed feature set {i(x)} could be infinite dimensional. The predicted class for a test instance x is determined as follows: For the binary classification data set described below, state whether it can be perfectly classified by a linear classier by choosing appropriate feature functions. Restrict the feature functions to polynomial expansions of the original attributes, e.g., 1(x) = x1x22 or 2(x) = x2x3x4, where x1, x2, x3, and x4 are part of the original attributes. If the answer is yes, write the mathematical expression for the linear classier f(x). Identify the feature functions i(x) as well as the parameters w in your expression. A data set with 4 continuous-valued features x1, x2, x3, and x4. The class label is +1 if the product of the x1 and x2 is greater than or equal to the product of x3 and x4; otherwise, it is 1.