Research Article
Privacy-Preserving Blockchain-Based Nonlinear SVM Classifier Training for Social Networks
Algorithm 2
Secure nonlinear SVM classifier training.
Input:: the split dataset learning rate , max iterations , and precision . | Output:: , . | (1) | initialize . | (2) | for from 1 to do | (3) | for from 1 to do | (4) | compute . | (5) | end for | (6) | end for | (7) | Whiledo | (8) | initialize . | (9) | compute . | (10) | for from 1 to do | (11) | for from 1 to do | (12) | compute . | (13) | locally compute . | (14) | end for | (15) | locally compute . | (16) | compute . | (17) | compute . | (18) | ifthen | (19) | compute . | (20) | compute . | (21) | end if | (22) | end for | (23) | update . | (24) | update . | (25) | compute . | (26) | . | (27) | end while | (28) | return . |
|