Data Balanced Algorithm Based on Generative Adversarial Network

I-Hsien Liu, Cheng En Hsieh, Wei Min Lin, Jung Shian Li, Chu Fen Li

研究成果: Conference contribution

摘要

In order to defend against malicious attacks, intrusion detection systems have introduced machine learning as a protection strategy. However, machine learning algorithms and datasets have a great influence on the effectiveness of the machine learning model. This study uses five algorithms which are Naïve Bayes, CNN, LSTM, BAT, and SVM to train the IDS machine learning model. We design a data-balanced method based on the GAN algorithm to improve the data imbalance problem of the IDS dataset.

原文English
主出版物標題Proceedings of the International Conference on Artificial Life and Robotics, ICAROB 2022
編輯Yingmin Jia, Takao Ito, Ju-Jang Lee
發行者ALife Robotics Corporation Ltd
頁面645-649
頁數5
ISBN(列印)9784990835071
出版狀態Published - 2022
事件27th International Conference on Artificial Life and Robotics, ICAROB 2022 - Virtual, Online
持續時間: 2022 1月 202022 1月 23

出版系列

名字Proceedings of International Conference on Artificial Life and Robotics
ISSN(電子)2435-9157

Conference

Conference27th International Conference on Artificial Life and Robotics, ICAROB 2022
城市Virtual, Online
期間22-01-2022-01-23

All Science Journal Classification (ASJC) codes

  • 人工智慧
  • 電腦視覺和模式識別
  • 硬體和架構
  • 資訊系統
  • 控制與系統工程
  • 電氣與電子工程
  • 建模與模擬

指紋

深入研究「Data Balanced Algorithm Based on Generative Adversarial Network」主題。共同形成了獨特的指紋。

引用此