Research Article
Corticomuscular Activity Modeling by Combining Partial Least Squares and Canonical Correlation Analysis
Algorithm 1
The combined PLS + CCA method.
Input: two data sets (with size ) and (with size ) | Output: corresponding LVs matrices , and | The First Step: | Solve the eigen decomposition problems: | and . | Determine and , the numbers of LVs extracted, corresponding to | the above two problems by the ratio of explained variance. | Determine the final number of LVs: . | Set . | Initialize both LVs matrices to be empty, that is, and . | while do | Set and to be the largest eigenvectors of the matrices | and , respectively. | Calculate the LVs as and . | Set and . | Deflate by subtracting the effects of the LV from the data space: | . | Deflate by subtracting the effects of the LV from the data space: | . | Let . | end while | The Second Step: | Solve the following eigen decomposition problems: | and | . | Set and to be the associated eigenvectors, respectively. | The recovered LVs and can be calculated by | and . |
|