VdaBSC: A Novel Vulnerability Detection Approach for Blockchain Smart Contract by Dynamic Analysis

Algorithm 2

Data Preprocessing Algorithm using RFDA and RRBN.

Require: Dataset (bytecode)

Ensure: Preprocessed dataset

    1: functionRANDOMFLIPPING()

    2:  Generate a binary mask with randomly selected flipped bits

    3:           Apply bitwise XOR

    4:  return

    5: end function

    6: functionREALTIMEBATCHNORMALIZATION()

    7:  Calculate the mean and standard deviation of the execution data

    8:           Apply normalization

    9:  return

 10: end function

 11: Load dataset

 12: Initiate the GRU memory model.

 13: Preprocessing Phase:

 14: for each contract in the dataset do

 15:  Extract bytecode from contract

 16:  Apply RandomFlipping function using eqn (1) to :

 17:  Extract execution data from contract

 18:  Apply RealTimeBatchNormalization function using eqn (2) to :

 19:  Replace the original bytecode and execution data in contract with and

 20: end for

 21: Return Preprocessed dataset