The invention discloses a shale gas reservoir identification method based on a support vector machine. The shale gas reservoir identification method is characterized by comprising the following steps of (a) synthesizing and classifying data of horizontal wells of which shale gas is exploited in the same block; (b) optimizing while-drilling parameters serving as a training sample by using regional data; (c) calculating (i) Q (/i) factors by using the while-drilling parameters; (d) optimally selecting model parameters of the support vector machine; (e) establishing a regional support vector machine model; and (f) identifying the reservoir of a horizontal well with exploiting shale gas by using the established model. The horizontal well in the step (f) serves as a prediction sample. According to known well logging interpretations and formation testing results, the optimized while-drilling parameters serve as judgment parameters, the model capable of effectively identifying the reservoir of the shale gas is analyzed and established through the support vector machine, while-drilling interpretation on the reservoir of the shale gas is realized, and the problems that shale gas reservoir logging interpretation methods are deficient, and lots of parameters are nonlinear are solved.