Learning edge weights in file co-occurrence graphs for malware detection

Learning edge weights in file co-occurrence graphs for malware detection
复制标题

学习文件共现图中的边缘权重以进行恶意软件检测

DOI:
10.1007/s10618-018-0593-7
复制
发表时间:
2018-10
影响因子:
4.8
通讯作者:
Guan Xiaohong
Guan Xiaohong
中科院分区:
计算机科学3区
文献类型:
--
作者:
Mao Weixuan;Cai Zhongmin;Zeng Bo;Guan Xiaohong

文献摘要

参考文献

相似文献

基于云的安全服务生成一种新型安全数据,该数据指示终端主机中可执行文件的出现。以安全数据为基础,文件共现图的半监督学习为恶意软件检测提供了新的视角。边缘权重可以量化同时出现的文件的标签(良性或恶意)的相关性,在此类技术中发挥着重要作用。虽然之前的工作采用启发式方法在文件共现图中定义边权重,但本文开发了一种新颖的框架,通过最小化图中隐含的调和属性下的误差来学习边权重。我们的方法被证明可以在训练实例图中实现独特的全局最优边缘权重。此外,利用同时出现的文件之间学习到的边缘权重,我们开发了一种基于图的半监督学习方法用于恶意软件检测。真实世界数据集包含来自 11,713,031 个终端主机的 12,469 个良性和 11,327 个恶意可执行文件,实验结果证明了我们方法的有效性。我们使用学习到的边缘权重的恶意软件检测方法明显优于使用常用启发式边缘权重的现有方法。
The cloud based security service generates a new type of security data, which indicates the occurrence of executable files in end hosts. With the basis of the security data, semi-supervised learning on file co-occurrence graph provides a novel perspective for malware detection. The edge weight, which quantifies the correlation of the labels (either benign or malicious) of co-occurred files, plays a significant role in such techniques. While previous work employed heuristic methods of defining edge weights in the file co-occurrence graph, this paper develops a novel framework for learning the edge weights via minimizing the error under the harmonic property which is implied from the graph. Our method is proven to achieve the unique global optimal edge weights in the graph of training instances. Furthermore, taking advantage of the learned edge weights between co-occurred files, we develop a graph based semi-supervised learning method for malware detection. Experimental results on a real-world dataset, which consists of 12,469 benign and 11,327 malicious executable files from 11,713,031 end hosts, demonstrate the efficacy of our method. Our malware detection approach with the learned edge weights significantly outperforms existing approaches with commonly used heuristic edge weights.
DOI: 10.1007/978-3-642-15880-3_42
发表时间: 2010-09
期刊: --
影响因子: --
作者:
Ming Ji;Yizhou Sun;Marina Danilevsky;Jiawei Han;Jing Gao
通讯作者: Ming Ji;Yizhou Sun;Marina Danilevsky;Jiawei Han;Jing Gao
DOI: 10.1016/j.knosys.2016.09.004
发表时间: 2016-11
期刊: Knowl. Based Syst.
影响因子: --
作者:
Ming Ni;Tao Li;Qianmu Li;Hong Zhang;Yanfang Ye
通讯作者: Ming Ni;Tao Li;Qianmu Li;Hong Zhang;Yanfang Ye
DOI: 10.1145/956863.956938
发表时间: 2003-11
期刊: --
影响因子: --
作者:
P. Calado;Marco Cristo;E. Moura;N. Ziviani;B. Ribeiro-Neto;Marcos André Gonçalves
通讯作者: P. Calado;Marco Cristo;E. Moura;N. Ziviani;B. Ribeiro-Neto;Marcos André Gonçalves
DOI: 10.1145/780542.780599
发表时间: 2003-06
期刊: J. Comput. Syst. Sci.
影响因子: --
作者:
Y. Azar;Edith Cohen;A. Fiat;Haim Kaplan;Harald Räcke
通讯作者: Y. Azar;Edith Cohen;A. Fiat;Haim Kaplan;Harald Räcke
DOI: --
发表时间: 2013-12
期刊: --
影响因子: --
作者:
Xiao-Ming Wu;Zhenguo Li;Shih-Fu Chang
通讯作者: Xiao-Ming Wu;Zhenguo Li;Shih-Fu Chang