The kernel methods are limited because the data of transformation can not be restored. For its characteristics of problem, a method based on Multi-kernel and Kernel ridge regression(KRR) is proposed. Firstly, the given data of homogeneous data is iterated by the kernel, and given data will express linearly in the ultra high-dimensional Euclidean space. Then, the unknown data can be linearly denoted by the given data. Thereafter, the unknown data can be regressed according to the KRR algorithm. Finally, the data reduction can be achieved. The two classes of data sets are selected as the experimental data. The results show that the proposed method has a great effect to restore unknown data, and it can be applied to other fields.