Comparisons among regression learning techniques results for model precision and accuracy.

All values of R presented as negative values are from [33] as such a negative R model would not yield useful MSE values. 1) MSE, labeled as RMSE. 2) accuracy was defined as “100 minus the average percentage difference between predicted and observed knockdown activities”. 3) a dataset of 702 siRNAs was used, not the 2431 dataset considered by the remainder of the table.