본문 바로가기

An Effective Intrusion Detection Classifier Using Long Short-Term Memory with Gradient Descent Optimization

문서 유형
연구출판물 · 학술대회 발표논문
등재
Scopus
피인용
Scopus 145
FWCI (Scopus)
11.8

저자

  1. 레티투흐엉제1저자School of Electrical and Computer Engineering, Pusan National University, Busan, South Korea
  2. 김지현School of Electrical and Computer Engineering, Pusan National University, Busan, South Korea
  3. 김호원School of Electrical and Computer Engineering, Pusan National University, Busan, South Korea

초록

Intrusion Detection System (IDS) is one of the important issues in network security. IDSs are built to detect both known and unknown malicious attacks. Several machine learning algorithms are used widely in IDS such as neural network, SVM, KNN etc. However, these algorithms have still some limitations such as high false positive and false alarm rate. In this paper, our contribution is to build a classifier of IDS following deep learning approach. We find the most suitable optimizer among six optimizes for Long Short-Term Memory Recurrent Neural Network (LSTM RNN) model applied in IDS. Through our experiments, we found that LSTM RNN model with Nadam optimizer outperforms to previous works. We demonstrate our approach is really efficiency to intrusion detection with accuracy is 97.54%, detection rate is 98.95%, and false alarm rate is reasonable with 9.98%.

키워드

Deep LearningGradient Descent OptimizationIntrusion Detection SystemLong Short-Term MemoryOptimizer

식별자

Scopus EID
2-s2.0-85018178265