期刊
DIGITAL INVESTIGATION
卷 22, 期 -, 页码 S106-S114出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.diin.2017.06.009
关键词
Digital forensics; Likelihood ratio; Marked point process; Event data; Density estimation; Time series
资金
- National Institute of Standards and Technology [70NANB15H176]
- Iowa State University [70NANB15H176]
- National Science Foundation [IIS-1320527]
- Google Faculty Award
- Direct For Computer & Info Scie & Enginr [1320527] Funding Source: National Science Foundation
- Div Of Information & Intelligent Systems [1320527] Funding Source: National Science Foundation
In this paper we investigate the application of score-based likelihood ratio techniques to the problem of detecting whether two time-stamped event streams were generated by the same source or by two different sources. We develop score functions for event data streams by building on ideas from the statistical modeling of marked point processes, focusing in particular on the coefficient of segregation and mingling index. The methodology is applied to a data set consisting of logs of computer activity over a 7-day period from 28 different individuals. Experimental results on known same-source and known different-source data sets indicate that the proposed scores have significant discriminative power in this context. The paper concludes with a discussion of the potential benefits and challenges that may arise from the application of statistical analysis to user-event data in digital forensics. (C) 2017 The Author(s). Published by Elsevier Ltd. on behalf of DFRWS.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据