BROWSE
Detail
Conference Paper
A Comparative Study on Feature Selection Methods for Anomaly Detection
Cited - time in 
Share 
Share
- Authors
- Eun Hye Kim, Seung Min Lee, Ki Hoon Kwon, Se Hun Kim
- Issue Date
- 2007-08
- Citation
- International Workshop on Information Security Applications (WISA) 2007, pp.1-11
- Language
- English
- Type
- Conference Paper
- Abstract
- For computationally efficient and effective IDS, it is essential to identify important input features. In this paper, we propose a hybrid feature selection method in which factor analysis based on PCA (Principal Components Analysis) is combined with optimized k-means clustering technique. Our approach hierarchically reduces the features which are redundant or contribute little to the detection process, thereby organizing a good subset of features critical to improve the performance of classifiers. Based on this result, we evaluate the performance of our hybrid feature selection method compared with feature selection algorithm using factor analysis. The experiment with KDD Cup 1999 data set shows several advantages in terms of computational complexity and our method achieves significant detection rate which shows possibility of detecting successfully attacks.