Title :
Research on the Intrusion Detection Based on the Improved BP Algorithm
Author :
Meng, Jianliang ; Wang, Linqian
Author_Institution :
Control & Comput. Eng, North China Electr. Power Univ., Baoding, China
Abstract :
BP learning algorithm for neural networks is an important method. However, due to its low detection efficiency and high error rate and undetected rate and so on, we propose an improved BP algorithm, which is based on Cauchy error estimator. Then, in order to make the experimental sample set better in line with the numerical characteristics of the network training, we filter the data set of KDDCup99. Finally, two methods are compared using Matlab programming. The results show that the improved method is better satisfactory to improve the detection performance of the system.
Keywords :
backpropagation; learning (artificial intelligence); neural nets; security of data; BP learning algorithm; Cauchy error estimator; KDDCup99; Matlab programming; intrusion detection; neural networks; Algorithm design and analysis; Biological neural networks; Decision trees; Educational institutions; Intrusion detection; BP algorithm; error estimator; intrusion detection; neural network;
Conference_Titel :
Computational and Information Sciences (ICCIS), 2012 Fourth International Conference on
Conference_Location :
Chongqing
Print_ISBN :
978-1-4673-2406-9
DOI :
10.1109/ICCIS.2012.258
