Skip to content

PIDS: A Behavioral Framework for Analysis and Detection of Network Printer Attacks #7

@palloc

Description

@palloc

公開日

2018-06-27

1. 概要

プリンターの悪性通信検知システムPIDSを提案。正常な通信と異常な通信を機械学習アルゴリズムで分類した。

2. 新規性・差分

  1. PIDSフレームワークの作成
  2. 汎用的で良質なプリンタの通信のデータセットの作成
  3. PRETの強化

3. 手法

image

feature extraction

  • パケットのサイズから28特徴
  • 時間から25特徴
  • TCPの情報から22特徴
    抽出している(分散とか標準偏差とか)。

4. 結果

5-cross validationで評価

Algorithm FPR TPR AUC Accuracy Length of time it takes to build the model (in seconds)
K-Means 0.313 0.761 0.721 75.46% 0.30
Decision Tree C4.5 0 1 0.999 99.95% 0.54
Naïve Bayes 0.311 0.507 0.816 50.74% 0.19
Bayesian Networks 0.005 0.995 1 99.55% 0.57
SVM 0.016 0.987 0.986 98.69% 1.72

各特徴のスコア
image

5. 議論

6. コメント

パケット系はdecision treeとか強いのわかる

論文情報・リンク

https://arxiv.org/pdf/1806.10642.pdf

Metadata

Metadata

Assignees

No one assigned

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions