Stratifying patients using fast multiple kernel learning framework: case studies of Alzheimer’s disease and cancers

摘要

The systematic pipeline of the proposed method. Step 1 is to collect data from different data sources. Step 2 is to pre-processing data, removing noisy data, rescaling data, normalizing data. Step 3 is to generate kernel matrices and then the kernel matrices are then combined in the Step 4. The number of dimensions is also optimised to obtain the most suitable numbers of dimensions and reduce the computation cost. The last step is to stratify patients using SVMs