Detecting Copy Directions among Programs Using Extreme Learning Machines

Algorithm 1

ELM training.

   Input: Training set ;

     Hidden node output function ;

     Number of hidden nodes ;

   Output: Weight vector ;

() for    to    do

()  Randomly generate hidden node parameters ;

() Calculate the hidden layer output matrix ;

() return the calculated output weight vector , where is the Moore-Penrose generalized by the inverse of matrix ;