College testing measurement is very important to choose talent. The quantification of tacit knowledge in test paper is the key to choose innovative talents. First of all, the traditional statistical methods are used to analyze the test results,and then to use support vector machine (SVM) method on the basis of it. It is concluded that the classification of different effects with the introduction of polynomial kernel function，radial basis kernel functions and takes both advantages of combined SVM kernel function training. and with different factors to show the quantitative comparative analysis of tacit knowledge. The experimental results show that the test results of evaluation should be based on the analysis of explicit and implicit knowledge analysis. Reasonable combination kernel function is applied to the pros and cons of the examination measurement is effective method.