Wireless Communications and Mobile Computing

Research Article

A Detection Method for Abnormal Transactions in E-Commerce Based on Extended Data Flow Conformance Checking

Algorithm 1

Outlier extraction.
Input:DPNE=(PN-DO, V, U, R, W, G, Φ), initial marking M0; Tx refers to the transition in Table 2, and x is the serial number, x∈[0, 31];
Output: event log L, the outlier sequences of L is stored in B;
1. L=∅, B=∅;
2. Let M0 be the root node, and mark it as “control-enabledness and data-enabledness”;
3. While “control-enabledness and data-enabledness” nodes exist Do
  Choose the “control-enabledness and data-enabledness” node as M';
  3.1 If ∃t ∈ T →M'[ct > and M'[dt > Then
       L=L∪{t};
     3.1.1 Ift =T6 or t=T30Then
            B[t]=1;
       Go to 3;
       End
     3.1.2 Else
    Selected the transition t and determine if it meets the guards;
       3.1.2.1 IfG(t)! = ∅ Then
      3.1.2.1.1 IfΦ(t) = ∅ Then
         3.1.2.1.1.1 Ift meets G(t) Then
                          B[t]=0;
                 End
         3.1.2.1.1.2 Else
                          B[t]=1;
                          Go to 3;
                 End
           End
            3.1.2.1.2 Else
       Get v(r) and v(w) of t as input, and according to the predetermined algorithm of current t, use it as Φ(t) to process data;
         3.1.2.1.2.1 Ift meets G(t) Then
                          B[t]=0;
                 End
         3.1.2.1.2.2 Else
                          B[t]=1;
                          Go to 3;
                 End
           End
         End
     3.1.2.2 Else
              B[t]=0
        Go to 3;
     End
       End
    End
  3.2 Else
       T=T-L; L=∅; B=∅;
      Go to 2;
  End
End