A Semi-Supervised Data Screening for Network Traffic Data Using Graph Min-Cuts

この論文をさがす

抄録

There are currently many projects aimed at devising efficient countermeasures against critical incidents occurring on the Internet through early detection. A nasty problem is hard-to-find accesses by well-analyzed malware whose packets make anomaly detection harder. In this paper, in order to find such accesses from raw data obtained by network monitoring, we propose an automatic data screening method using graph-based semi-supervised learning (Blum and Chawla, 2001) and show its effectiveness in experiments on darknet traffic.

There are currently many projects aimed at devising efficient countermeasures against critical incidents occurring on the Internet through early detection. A nasty problem is hard-to-find accesses by well-analyzed malware whose packets make anomaly detection harder. In this paper, in order to find such accesses from raw data obtained by network monitoring, we propose an automatic data screening method using graph-based semi-supervised learning (Blum and Chawla, 2001) and show its effectiveness in experiments on darknet traffic.

収録刊行物

関連プロジェクト

もっと見る

詳細情報

  • CRID
    1050282812883882240
  • NII論文ID
    170000148131
  • NII書誌ID
    AA11464803
  • ISSN
    18827780
  • Web Site
    http://id.nii.ac.jp/1001/00174172/
  • 本文言語コード
    en
  • 資料種別
    article
  • データソース種別
    • IRDB
    • CiNii Articles
    • KAKEN

問題の指摘

ページトップへ