Random Forest Bagging and X-Means Clustered Antipattern Detection from SQL Query Log for Accessing Secure Mobile Data

Algorithm 1

Random Forest Bagging X-means Clustering.

Input: Number of patterns (i.e. queries).’.

Output: Accurate and timely anti-pattern identification.

1: Initialize ‘ number of weak X-means clusters’.

2: Begin.

3: For each query pattern’.

4: Randomly assign weight to cluster ‘.

5: Measure similarity between cluster weight ‘ and input patterns ‘.

6: If similarity coefficient ‘ is 1, then ‘ and ‘ are more similar.

7: Else ‘ and ‘ are dissimilar.

8: End if.

9: Group similar patterns ‘ into cluster ‘.

10: If ‘ not grouped into cluster apply Bayesian information criterion to group patterns 11: Else.

12: Combine all the weak clusters results using (5).

13: Apply voting using (6).

14: Return (anti-patterns).

15: End if.

16: End for.

17: End.