Intrusion Detection Based
on Convolutional Neural Network
in Complex Network Environment
Yunfeng Zhou
1 , Xuezhang Zhu
1 , Su Hu
1(B) , Di Lin
1 , and Yuan Gao
1,2
1 University of Electronic Science and Technology of China, Sichuan 611731, China
Husu@uestc.edu.cn
2 Academy of Military Science of PLA, Beijing 100090, China
Abstract. In this paper, we propose an intrusion detection method
based on a convolutional neural network (CNN) for intrusion detection
systems. In designing a deep intrusion detection model, mainstream deep
learning techniques such as dropout, Adam, and Softmax classifier are
used, respectively. Firstly, plain text data is dimensionally corrected and
converted into grayscale images. Secondly, the CNN model obtains the
feature map by learning features. Finally, the feature map is input to
the Softmax classifier to obtain the detection results. The method is
implemented on TensorFlow and tested on KDDCUP’99 data set. The
results show that the model proposed can obtain high detection accuracy rapidly and satisfy the real-time detection requirements of complex
network systems.
Keywords: Intrusion detection · Convolutional neural network · Deep
learning · Complex network
1 Introduction
In recent years, due to the rapid development of the Internet and its wide application in various fields of society, the problem of network security has become
increasingly serious. Intrusion detection systems play a critical role in the maintenance of network security systems because it can generate an alarm in the
event of an attack when monitoring network traffic [1–4]. The essence of intrusion detection can be summarized as the classification of normal data and attack
data [5–7], and machine learning algorithm is one of the most effective algorithms
in classification problems [8].
Classification algorithms in machine learning include support vector
machines (SVM), K-means clustering, decision tree, etc. SVM is sensitive to
parameter setting when dealing with large sample data. K-means clustering is
an unsupervised algorithm, so the known labels in the sample cannot be used.
c
Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 229–238, 2020.
https://doi.org/10.1007/978-981-15-0187-6_26
on Convolutional Neural Network
in Complex Network Environment
Yunfeng Zhou
1 , Xuezhang Zhu
1 , Su Hu
1(B) , Di Lin
1 , and Yuan Gao
1,2
1 University of Electronic Science and Technology of China, Sichuan 611731, China
Husu@uestc.edu.cn
2 Academy of Military Science of PLA, Beijing 100090, China
Abstract. In this paper, we propose an intrusion detection method
based on a convolutional neural network (CNN) for intrusion detection
systems. In designing a deep intrusion detection model, mainstream deep
learning techniques such as dropout, Adam, and Softmax classifier are
used, respectively. Firstly, plain text data is dimensionally corrected and
converted into grayscale images. Secondly, the CNN model obtains the
feature map by learning features. Finally, the feature map is input to
the Softmax classifier to obtain the detection results. The method is
implemented on TensorFlow and tested on KDDCUP’99 data set. The
results show that the model proposed can obtain high detection accuracy rapidly and satisfy the real-time detection requirements of complex
network systems.
Keywords: Intrusion detection · Convolutional neural network · Deep
learning · Complex network
1 Introduction
In recent years, due to the rapid development of the Internet and its wide application in various fields of society, the problem of network security has become
increasingly serious. Intrusion detection systems play a critical role in the maintenance of network security systems because it can generate an alarm in the
event of an attack when monitoring network traffic [1–4]. The essence of intrusion detection can be summarized as the classification of normal data and attack
data [5–7], and machine learning algorithm is one of the most effective algorithms
in classification problems [8].
Classification algorithms in machine learning include support vector
machines (SVM), K-means clustering, decision tree, etc. SVM is sensitive to
parameter setting when dealing with large sample data. K-means clustering is
an unsupervised algorithm, so the known labels in the sample cannot be used.
c
Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 229–238, 2020.
https://doi.org/10.1007/978-981-15-0187-6_26