An exploratory study into deception detection in text-based computer-mediated communication

Zhou, Lina; Twitchell, Douglas P.; Qin, Tiantian; Burgoon, Judee K.; Nunamaker, Jay F.

doi:10.1109/hicss.2003.1173793

Cited by 113 publications

(107 citation statements)

References 18 publications

Supporting

Mentioning

100

Contrasting

Unclassified

Order By: Relevance

“…Nevertheless, just the opposite appears to be true which, yet again, points to the psychological and cognitive complexity of constructing high-stakes lies [31,10]. Additionally, less variables were used by non-suicidal writers, which is congruent with Zhou et al [32] and Zhou et al's [33] argument that deceptive accounts are less diverse in content.…”

Section: Themes In Genuine and Simulated Suicide Notesmentioning

confidence: 78%

Genuine and simulated suicide notes: An analysis of content

Ιωάννου

Debowska

2014

Forensic Science International

View full text Add to dashboard Cite

The present study examined genuine and simulated suicide notes aiming to identify the measures of content that best differentiate between the two. Thirty-three genuine and thirty-three simulated suicide notes were content-analysed and data subjected to Smallest Space Analysis (SSA), a Multidimensional Scaling Procedure. The core of all suicide notes was discovered to be constructed with the use of three variables: expressions of love, positive construction of partner and apologies. Furthermore, four different genuine suicide note themes ('planned escape', 'negative affect and self-mitigation', 'positive affect and failed relationship', 'lack of selfacceptance') and three simulated suicide note themes ('escape', 'positive affect and self-blame', 'purposeless life') were identified revealing that authentic suicide note themes were more internally consistent and clearer to interpret.Keywords: suicide; suicide notes; genuine suicide notes; simulated suicide notes; MDS analysis 2 Suicide note writing has an important social dimension [1]. Suicide notes are a window to the manner in which suicides construct themselves, others and their surrounding reality and consequently -to the reasons of death. This is reflected in a study [2] inquiring into the discursive self-presentation of suicidal individuals in online interactions. The researchers reveal how their subjects' ability to think prospectively is disturbed. Suicides may be motivated to leave a note in order to have their needs understood by others as suicide notes are always aimed at audience and hence they create a social act -they transport meanings and construct suicide's reality. As [3] report, suicide notes constitute a peculiar self-continuation and enable the deceased to exert some influence even after death. Even more, they hold the power to connect the writers to their close ones. It, therefore, seems important to study the content of suicide notes in order to understand suicidal individuals.

show abstract

Section: Themes In Genuine and Simulated Suicide Notesmentioning

confidence: 78%

Genuine and simulated suicide notes: An analysis of content

Ιωάννου

Debowska

2014

Forensic Science International

View full text Add to dashboard Cite

show abstract

“…Zhou et al (2004b) concluded that individual references and group references are important features in machine learning models that detect deception in computer-mediated messages. Such messages contain less self-references and more group references (Zhou et al, 2004a(Zhou et al, , 2003. However, in computer-mediated messages regarding audit-related questions, the deceivers used more first person singular pronouns ).…”

Section: Grammatical Featuresmentioning

confidence: 99%

“…Therefore, the complexity of the texts lies in the words and sentences. For all complexity features, the assumption holds that longer words and sentences are an indication of a more complex text (Zhou et al, 2003).…”

Section: Complexity Featuresmentioning

confidence: 99%

“…The text mining model of Afroz et al (2012) that detects deceptive adversarial documents includes the number of long words, the number of syllables per word, the average number of words per sentence and the number of short and long sentences. In computer-mediated communication, the deceptive messages contained less punctuation, fewer long sentences and fewer syllables per word (Zhou et al, 2003). Therefore, Zhou et al (2004b) used these features in their machine learning model to detect such messages.…”

Section: Complexity Featuresmentioning

confidence: 99%

See 1 more Smart Citation

Text mining to detect indications of fraud in annual reports worldwide

Fissette¹

View full text Add to dashboard Cite

“…Over the past two and a half years, the Center for the Management of Information (CMI) at the University of Arizona has conducted over a dozen experiments to study deception with over 2000 subjects [2][3][4][5]. These experiments have been instrumental in creating an understanding of the factors influencing deception, and have guided the building of automated tools for enhancing deception detection and the creation of training for security personnel [3,6].…”

Section: Deception Intent and Behaviormentioning

confidence: 99%