This paper proposes a new approach to feature selection based on a statistical feature mining technique for sequence and tree kernels. Since natural language data take discrete structures, convolution kernels, such as sequence and tree kernels, are advantageous for both the concept and accuracy of many natural language processing tasks. However, experiments have shown that the best results can only be achieved when limited small sub-structures are dealt with by these kernels. This paper discusses this issue of convolution kernels and then proposes a statistical feature selection that enable us to use larger sub-structures effectively. The proposed method, in order to execute efficiently, can be embedded into an original kernel calculation process by using sub-structure mining algorithms. Experiments on real NLP tasks confirm the problem in the conventional method and compare the performance of a conventional method to that of the proposed method.