Abstract
Utilizing the feature selection in the intrusion detection can delete the redundant features on the base of protecting the integrity of original data and improve the detection speed of the system efficiently. This paper proposes a new feature selection method that is based on KNN and Tabu search algorithm. The experiment result shows that this method can remove the redundant features, and reduce the time of feature selection. This method not only can guarantee the correct rate of detection but also improve the detection speed efficiently.
| Original language | English |
|---|---|
| Pages (from-to) | 1628-1632 |
| Number of pages | 5 |
| Journal | Tien Tzu Hsueh Pao/Acta Electronica Sinica |
| Volume | 37 |
| Issue number | 7 |
| Publication status | Published - Jul 2009 |
Keywords
- Feature relevance
- Feature selection
- Intrusion detection
- Tabu search