Research Article

Identifying Fake Accounts on Social Networks Based on Graph Analysis and Classification Algorithms

Box 1

Input:
Graph G (N, E) with labeled node
Output:
List of the fake node (account)
Procedure:
A adjacency matrix (G)
For all similarities
SM calculate similarity matrix
SM feature extraction by PCA Algorithm on (SM)
End
SM select ten columns in all similarity matrix (SM)
SM balance SMOTE on (SM)
Training Classifier
Return  List of fake nodes
Send alarm for fake accounts friends.