Multi-Parameter Sensitivity Analysis of a Bayesian Network from a Digital Forensic Investigation

Richard Overill; Echo P. Zhang; Kam-Pui Chow

Multi-Parameter Sensitivity Analysis of a Bayesian Network from a Digital Forensic Investigation

Richard Overill, Echo P. Zhang, Kam-Pui Chow

Research output: Chapter in Book/Report/Conference proceeding › Conference paper

Abstract

A multi-parameter sensitivity analysis of a Bayesian network (BN) used. in the digital forensic investigation of the Yahoo! email case has been performed using the principle of 'steepest gradient' in the parameter space of the conditional probabilities. This procedure delivers a more reliable result for the dependence of the posterior probability of the BN on the values used to populate the conditional probability tables (CPTs) of the BN. As such, this work extends our previous studies using single-parameter sensitivity analyses of BNs, with the overall aim of more deeply understanding the indicative use of BNs within the digital forensic and prosecutorial processes. In particular, we find that while our previous conclusions regarding the Yahoo! email case are generally validated by the results of the multi-parameter sensitivity analysis, the utility of performing the latter analysis as a means of verifying the structure and form adopted for the Bayesian network should not be underestimated.

Original language	English
Title of host publication	Proceedings of the 7th Annual ADFSL Conference on Digital Forensics, Security and Law
Publisher	Association of Digital Forensics, Security, and Law
Pages	129-+
Volume	N/A
Edition	N/A
Publication status	Published - 2012

Cite this

@inbook{2431c29b70c84b7795141551b2e8b59a,

title = "Multi-Parameter Sensitivity Analysis of a Bayesian Network from a Digital Forensic Investigation",

abstract = "A multi-parameter sensitivity analysis of a Bayesian network (BN) used. in the digital forensic investigation of the Yahoo! email case has been performed using the principle of 'steepest gradient' in the parameter space of the conditional probabilities. This procedure delivers a more reliable result for the dependence of the posterior probability of the BN on the values used to populate the conditional probability tables (CPTs) of the BN. As such, this work extends our previous studies using single-parameter sensitivity analyses of BNs, with the overall aim of more deeply understanding the indicative use of BNs within the digital forensic and prosecutorial processes. In particular, we find that while our previous conclusions regarding the Yahoo! email case are generally validated by the results of the multi-parameter sensitivity analysis, the utility of performing the latter analysis as a means of verifying the structure and form adopted for the Bayesian network should not be underestimated.",

author = "Richard Overill and Zhang, {Echo P.} and Kam-Pui Chow",

year = "2012",

language = "English",

volume = "N/A",

pages = "129--+",

booktitle = "Proceedings of the 7th Annual ADFSL Conference on Digital Forensics, Security and Law",

publisher = "Association of Digital Forensics, Security, and Law",

edition = "N/A",

}

Multi-Parameter Sensitivity Analysis of a Bayesian Network from a Digital Forensic Investigation. / Overill, Richard; Zhang, Echo P.; Chow, Kam-Pui.
Proceedings of the 7th Annual ADFSL Conference on Digital Forensics, Security and Law. Vol. N/A N/A. ed. Association of Digital Forensics, Security, and Law, 2012. p. 129-+.

Research output: Chapter in Book/Report/Conference proceeding › Conference paper

TY - CHAP

T1 - Multi-Parameter Sensitivity Analysis of a Bayesian Network from a Digital Forensic Investigation

AU - Overill, Richard

AU - Zhang, Echo P.

AU - Chow, Kam-Pui

PY - 2012

Y1 - 2012

N2 - A multi-parameter sensitivity analysis of a Bayesian network (BN) used. in the digital forensic investigation of the Yahoo! email case has been performed using the principle of 'steepest gradient' in the parameter space of the conditional probabilities. This procedure delivers a more reliable result for the dependence of the posterior probability of the BN on the values used to populate the conditional probability tables (CPTs) of the BN. As such, this work extends our previous studies using single-parameter sensitivity analyses of BNs, with the overall aim of more deeply understanding the indicative use of BNs within the digital forensic and prosecutorial processes. In particular, we find that while our previous conclusions regarding the Yahoo! email case are generally validated by the results of the multi-parameter sensitivity analysis, the utility of performing the latter analysis as a means of verifying the structure and form adopted for the Bayesian network should not be underestimated.

AB - A multi-parameter sensitivity analysis of a Bayesian network (BN) used. in the digital forensic investigation of the Yahoo! email case has been performed using the principle of 'steepest gradient' in the parameter space of the conditional probabilities. This procedure delivers a more reliable result for the dependence of the posterior probability of the BN on the values used to populate the conditional probability tables (CPTs) of the BN. As such, this work extends our previous studies using single-parameter sensitivity analyses of BNs, with the overall aim of more deeply understanding the indicative use of BNs within the digital forensic and prosecutorial processes. In particular, we find that while our previous conclusions regarding the Yahoo! email case are generally validated by the results of the multi-parameter sensitivity analysis, the utility of performing the latter analysis as a means of verifying the structure and form adopted for the Bayesian network should not be underestimated.

M3 - Conference paper

VL - N/A

SP - 129-+

BT - Proceedings of the 7th Annual ADFSL Conference on Digital Forensics, Security and Law

PB - Association of Digital Forensics, Security, and Law

ER -

Multi-Parameter Sensitivity Analysis of a Bayesian Network from a Digital Forensic Investigation

Abstract

Fingerprint

Cite this