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

Algorithm 2

Abnormal order detection.

Input:

1. The outlier sequence set is composed of the transition set in Algorithm 1;

2. The set of user’s history transaction behavior patterns and the normal patterns , and its transition sets is represented by ;

3. The weight set of each activity is sequence , where ;

4. The weight of each normal pattern is sequence , where ;

5. is the set of all normal behavior patterns ;

6. is the set of matching results of each activity;

Output: the path to be selected by the system.

1. Calculate the matching results between and and store it in , initial ;

2. While the elements of the set N are not all traversed Do

   select from ;

  While the elements of the set are not all traversed Do

   select from ;

  IfThen

   ;

  Else

   ;

  End

End

3. Calculate each matching results under the current normal behavior patterns by formula (2);

4. Calculate the final anomaly detection result by formula (3);

 4.1 IfThen

  ;

 4.2 Else

  ;