Research Article
[Retracted] Research on Boruta-ET-Based Anomalous Traffic Detection Model
| | | Input: Train set | | Output: Extreme Trees | (1) | for i = 1 to M do | (2) | Generating decision trees, | (3) | Return Extreme Trees T | (4) | end for | | Build_an_extra_tree(D) | | Input: Train data | | Output: Decision Tree t | (1) | if or all candidate attributes in D are constant or output variables in D are constant then | (2) | Return a leaf node | (3) | else | (4) | Randomly select K attributes from all candidate attributes | (5) | Generate K split thresholds , Among them | (6) | According to , Selecting the best test split threshold | (7) | According to test split thresholds , Divide the sample set D into two sub-sample sets and | (8) | Construct a left subtree and a right subtree using subsets and respectively | (9) | Create a tree node based on , with and as its left and right subtrees respectively, and return a decision tree t | (10) | end if | | | | Input: Train data , Attributes a | | Output: Divided attributes | (1) | Calculate the minimum and maximum values of attribute a in the training set D, denoted respectively as and | (2) | Select a random splitting attribute from | (3) | Return to Split attributes |
|