| Popis: |
Aiming at the classification detection problem of intrusion detection system, this paper proposes a classification algorithm based on approximate information entropy and random forest. First, the training data is subjected to dimensionality reduction and noise reduction by approximating information entropy, and redundant attributes are deleted. Secondly, the processed data is classified using the random forest classification algorithm. Finally, in order to verify the effectiveness of the algorithm, the improved method was tested using the KDD-CUP99 data set. The experimental results show that the data with approximate information entropy dimensionality reduction and noise reduction can effectively reduce the time complexity of classification and improve the classification accuracy. |
