ナイーブベイズを用いたDrive-by-Download攻撃予測の評価

Search this article

Abstract

Recenty, a damage by drive-by-download attacks is increasing, in which after the user browses a compromised website, a malware is automatically downloaded into her/his computer using exploited vulnerabilities. Although there are several countermeasures against drive-by-download attacks, effective one does not have been proposed so far. In this paper, we focus on attack prediciton using exploited vulnerability group. Exploited vulnerabilities can be partially related, hence we classify vulnerabilities into same group by similary charactersitics and predict attacks from such vulnerability groups. To the best of our knowlege, our proposed method is the first one to predict attacks using such grouping method. We use two kinds of machine learning algorithms, K-means++ for grouping vulnerabilities and Naive Bayes for prediting attacks. The accuracy of prediction are improved from the results of our evaluation. We use D3M dataset from 2010 to 2013.

Journal

  • IPSJ SIG Notes

    IPSJ SIG Notes 2014 (19), 1-6, 2014-05-15

    Information Processing Society of Japan (IPSJ)

Details 詳細情報について

  • CRID
    1570572702879978752
  • NII Article ID
    110009771656
  • NII Book ID
    AA11235941
  • ISSN
    09196072
  • Text Lang
    ja
  • Data Source
    • CiNii Articles

Report a problem

Back to top