Figure 8
AdaBoost
1. Initialize the observation weights
2. For m = 1 to M repeat steps :
(a) Fit a classifier to the training data using weights.
(b) Compute weighted error of newest tree
(c) Compute
(d) Update weights for
And renormalize to to sum to 1.
3. Output
Therefore, our proposed work is efficient in analyzing the network traffic as well as in recognizing and removing the bot net attack which is influenced by a botmaster .