A likelihood ratio-based forensic text comparison in predatory chatlog messages

Shunichi Ishihara

A likelihood ratio-based forensic text comparison in predatory chatlog messages

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

An experiment in Forensic Text Comparison (FTC) within the Likelihood Ratio (LR) framework is described, which determines the strength of authorship attribution evidence from chatlog messages using so-called lexical features. More specifically, in this study I will investigate 1) the degree of evidential strength (or LR) that can be obtained from chatlog messages and 2) how the performance of the FTC system and the magnitudes of the LRs are influenced by the sample size for modelling. The performance of the system is assessed using the log-LR cost (Cllr) and the magnitudes of the obtained LRs are visually presented as Tippett plots. It is demonstrated in this study that you can use the lexical features within the LR framework to discriminate same-author and different-author chatlog messages.

Original language	English
Title of host publication	Selected Papers from the 44th Conference of the Australian Linguistic Society, 2013
Editors	Lauren Gawne and Jill Vaughan
Place of Publication	Melbourne
Publisher	Australian Linguistic Society
Pages	41-57
Edition	Peer reviewed
Publication status	Published - 2014
Event	the 44th Conference of the Australian Linguistic Society - Melbourne, Australia, Australia Duration: 1 Jan 2014 → …

Conference

Conference	the 44th Conference of the Australian Linguistic Society
Country/Territory	Australia
Period	1/01/14 → …
Other	1-4 October 2013

Cite this

@inproceedings{6edd59f4046d4caeb2d4c9c17833c008,

title = "A likelihood ratio-based forensic text comparison in predatory chatlog messages",

abstract = "An experiment in Forensic Text Comparison (FTC) within the Likelihood Ratio (LR) framework is described, which determines the strength of authorship attribution evidence from chatlog messages using so-called lexical features. More specifically, in this study I will investigate 1) the degree of evidential strength (or LR) that can be obtained from chatlog messages and 2) how the performance of the FTC system and the magnitudes of the LRs are influenced by the sample size for modelling. The performance of the system is assessed using the log-LR cost (Cllr) and the magnitudes of the obtained LRs are visually presented as Tippett plots. It is demonstrated in this study that you can use the lexical features within the LR framework to discriminate same-author and different-author chatlog messages.",

author = "Shunichi Ishihara",

year = "2014",

language = "English",

pages = "41--57",

editor = "{Lauren Gawne and Jill Vaughan}",

booktitle = "Selected Papers from the 44th Conference of the Australian Linguistic Society, 2013",

publisher = "Australian Linguistic Society",

edition = "Peer reviewed",

note = "the 44th Conference of the Australian Linguistic Society ; Conference date: 01-01-2014",

}

A likelihood ratio-based forensic text comparison in predatory chatlog messages. / Ishihara, Shunichi.
Selected Papers from the 44th Conference of the Australian Linguistic Society, 2013. ed. / Lauren Gawne and Jill Vaughan. Peer reviewed. ed. Melbourne: Australian Linguistic Society, 2014. p. 41-57.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A likelihood ratio-based forensic text comparison in predatory chatlog messages

AU - Ishihara, Shunichi

PY - 2014

Y1 - 2014

N2 - An experiment in Forensic Text Comparison (FTC) within the Likelihood Ratio (LR) framework is described, which determines the strength of authorship attribution evidence from chatlog messages using so-called lexical features. More specifically, in this study I will investigate 1) the degree of evidential strength (or LR) that can be obtained from chatlog messages and 2) how the performance of the FTC system and the magnitudes of the LRs are influenced by the sample size for modelling. The performance of the system is assessed using the log-LR cost (Cllr) and the magnitudes of the obtained LRs are visually presented as Tippett plots. It is demonstrated in this study that you can use the lexical features within the LR framework to discriminate same-author and different-author chatlog messages.

AB - An experiment in Forensic Text Comparison (FTC) within the Likelihood Ratio (LR) framework is described, which determines the strength of authorship attribution evidence from chatlog messages using so-called lexical features. More specifically, in this study I will investigate 1) the degree of evidential strength (or LR) that can be obtained from chatlog messages and 2) how the performance of the FTC system and the magnitudes of the LRs are influenced by the sample size for modelling. The performance of the system is assessed using the log-LR cost (Cllr) and the magnitudes of the obtained LRs are visually presented as Tippett plots. It is demonstrated in this study that you can use the lexical features within the LR framework to discriminate same-author and different-author chatlog messages.

M3 - Conference contribution

SP - 41

EP - 57

BT - Selected Papers from the 44th Conference of the Australian Linguistic Society, 2013

A2 - null, Lauren Gawne and Jill Vaughan

PB - Australian Linguistic Society

CY - Melbourne

T2 - the 44th Conference of the Australian Linguistic Society

Y2 - 1 January 2014

ER -

A likelihood ratio-based forensic text comparison in predatory chatlog messages

Abstract

Conference

Fingerprint

Cite this